P 079_Proposal_to_ONR_for_Research_on_Guidance_to_Planners_of_ANTACCS_Attachement_II 079 Proposal To ONR For Research On Guidance Planners Of ANTACCS Attachement II
P-079_Proposal_to_ONR_for_Research_on_Guidance_to_Planners_of_ANTACCS_Attachement_II P-079_Proposal_to_ONR_for_Research_on_Guidance_to_Planners_of_ANTACCS_Attachement_II
User Manual: P-079_Proposal_to_ONR_for_Research_on_Guidance_to_Planners_of_ANTACCS_Attachement_II
Open the PDF directly: View PDF 
.
Page Count: 172
| Download | |
| Open PDF In Browser | View PDF |
r---··-------·---·--·--·-·--· -._ .. -., ----.-..---.---,---------- - - - - - - - - - -.. -. ------- .... '-;--- ...- -
.I
I
P-079
PROPOSAL TO
OFfiCE OF NAVAL RESEARCH
FOR RESEARCH ON
GUIDANCE TO PLANNERS
OF ADVANCED NAVY
COMMAND AND CONTROL
SYSTEMS
ATTACHMENT
f
I.
;
I
;
I
__________.-_______________._.____________.____._______. _ ..____._____________._____ . _ . __.__._.____J
P9958-079
TABLE OF CONTENTS
LIST.OF ILLUSTRATIONS
SECTION 1
SECTION 2
SECTION 3
INTRODUCTION
SUMMARY
3.3
RESEARCH METHODOLOGY & TECHNICAL APPROACH
Subtask Interfaces
Technology Research
Applicability Criteria
3-8
3.4
Study Integration
3-10
3.5
System Design Methodology
3-12
COMPUTER ORGANIZATION
Introduction and Definition
4-1
3. 1
3.2
SECTION 4
4. 1
4.2
3-1
3-5
Multi-Computers and Multi-Processing
Highly Parallel Machines
4-2
4-5
5. 1
Stored Logic and Micro-Programmed Computers
Analog/Digital Hybrid Techniques
INPUT/OUTPUT AND DISPLAYS
Definitions and Functions
5.2
5.3
Input/Output Buffering and Interrupt Handling
Input/Output Devices
5-3
5.4
Console Display and- Interrogation
Group Displays
Communications Data Handling
5-29
HARDWARE TECHNIQUES
General
Basic Components
6-1
4.3
4.4
4.5
SECTION 5
5.S
5.6
SECTION 6
6. 1
6.2
6.3
6.4
SECTION 7
7. 1
7.2
7.3
4-3
4-7
5-1
5-9
5-12
5-36
6-2
Integrated Circuits and Batch Fabrication
Techniques
6-3
Reliability, Maintainabf lity, and
Serv iceabi 1 i ty
6-5
MEMORY TECHNIQUES
Operational Considerations
Memory Hardware Techniques
7-4
Mass Memory
7-9
7-1
•I
•I
P9958-079
TABLE OF CONTENTS (CONTINUED)
SECTION 8
PROGRAMMI NG
8. 1
Assemblers and Compilers
8-1
8.2
8.3
Executive and Master Control
8-3
Programming Management
8-11
SECTION 9
ADVANCED USAGE TECHNIQUES
9. 1
Learning and Self-Diagnosing Techniques
9-1
9.2
Heuristic Problem Solving
Language Translation
Applications in Future Command and Control
Systems
COMMENTS ON TWO NAVY SYSTEMS
Functions and Requirements for Naval Tactical
Data Systems
National Emergency Command Post Afloat
(NECPA)
9-2
9.3
9.4
SECTION 10
10. 1
10.2
SECTION I 1
J 1. 1
11.2
11.3
11.4
SECTION 12
PROJECT PLAN
Work Statement and Delivery Schedule
Work Plan, Milestones and Schedules
Project Control System
The Project Team
9-3
9-4
10-1
10-3
11- 1
11- 3
11-6
11-6
CONTRACTOR QUALIFICATIONS
12. 1
Informatics tnc. and Hobbs Associates
12- 1
12.2
Team Qualifications
12-2
12.3
12-6
12.4
Individual Qualifications
Bibliography of Team Members
12.5
Resumes of Personnel
12- 21
12-14
•I
•I
,.---_._-------------_._-----_._-------,.-----_ .. _ -
P9958-079
LIST OF ILLUSTRATIONS
Figure
No.
Ti tle
Page No.
Computer Requirements Related to System
Factors
3-2
3-2
Man-Machine Coordination
3-4
3-3
Current and Projected Technology Research
3-7
3-4
Basic Ways of Handling Communication Within
the System
3-11
3-5
Queuing Theory Application
3-13
3-6
Computer/On-Line Device Configurations
3-14
3-7
Summary of Equipment and Design Trade-offs
3-15
3-8
General Implementation Procedure for On-Line
Systems
3- 16
3-9
Information System Design Methodology
3- 18
3-10
Queuing Analysis Example
3-20
4-1
Functional Assignments Between Hierarchy
and Distributed Concepts
4-4
Interrupt Handling in Conventional and
On-Line Processors
5-7
Interrupt Characteristics for a Selected
History of Computers
5-8
3-1
5-}
5-2
5-3
Candidate Console Characteristics
5-14
5-15
5-16
5-4
Console Applications and Requirements
5-17
5-5
Computer On-line Oevice Configurations
5-19
5 ... 6
Probability of n Consoles of Ten Requiring
Service at the Same Time
5-26
Comparative Characteristics of Selected
Display Systems
5-33
5-8
Schematic of Interface Functions
5-39
5 ... 9
Typical Saturn V Stage I I Checkout
Conf i gurat ion
5-40
Typical PMR Rea1 Time Data Handling Buffer
Con f i gu rat I on
5-40
5-7
5-10
•
I
I
•
P9958-079
LIST OF ILLUSTRATIONS (CONTINUED)
Figure No.
TI tIe
Page No.
7-1
Summary of Characteristics of Mass Memories
7-10
7-2
Advantages and Disadvantages of Types of
Ma s s Memo r i e s
7-11
8-1
Executive Control System Hierarchy
8-10
8-2
Statistics for Implementing One Command and
Cont ro I Sys tern
8-13
11-4
11-7
11- 1
Project Schedule
1t .. 2
Proj ec t Organ i za t i on
Contact of Team Members with Large-Scale
Hi 1 i tary Systems
12-4
Summary of Qualifications of Project Team
Members
12-7
12- 1
12-2
•I
•I
- - - - - - - - - - - - - - - - - - _.._..
---
- .-._--_._--_.-
P9958-079
Page 1-1
1.
INTRODUCTION
This is a proposal by Informatics Inc. to supply research
services to the Office of Naval Research to produce documented information which will provide guidance to planners of advanced Navy
command and control systems.
The Project Plan calls for the consulting
services of one man year from Hobbs Associates in the area of computer
hardware.
The proposal is for two alternative contracts, one for
Subtask 2 only (Current and Projected Technology) and the other for
Subtasks 2 and 3 (Current and Projected Technology, Study Integration
and Research Methodology).
If Informatics Inc. is considered for
only Subtask 2, the comments made herein regarding Subtask 3 wi 11
convey the fact that there is an appreciation by the proposer of
Subtask 3 which will be
import~nt
In the contract effort.
This is a lengthy proposal despite the fact that contributors
were told to write 11briefly but comprehensivel y ' 1 on the subject.
However, we have organized the proposal in such a way that persons
evaluating It can easily and quickly go to the subject of their
Interest.
In addition to the outline provided by the table of
contents, Section 2 provides a summary of the entire proposal.
Section 3 is the Research Methodology and Technical Approach description which covers what wl11 be done on the project and al I of
the technical areas.
technical areas.
Sections 4 through 10 cover the data processing
Section 11 is the Project Plan which describes what,
how, and by whom the work is to be done.
Section 12 is the contractor
qua 1 if i ca t ions.
One of the reasons why this is a lengthy proposal is that the
subject matter is very broad.
Future Navy command and control systems
wilt use nearly every existing or conceived data processing technology
or technique.
The requirements of speed, storage and system
•I
•I
-"._--------"--"-------------
P9958-079
Page 1-2
organization are very broad.
They encompass, at least potentially,
the entire spectrum of developments for future systems.
There Is
hardly a new technology which can be unqualifiedly said to be inappropriate for use in the operational
10 years hence.
c~nd
and control systems of
Because the requirements are broad in scope, and consequently
the technical areas likewise are broad in scope, so must the contractor
and the project personnel have wide experience. Obviously this
experience must cover the entire data processing field so that the
key proposed project members must themselves have wide experience in
the field. On the other hand, they must be incisive researchers and
not people who develop general or platitudinous approaches.
We regard this work as a research effort in the strict sense.
Our approach, whi Ie developed from the point of view of the pragmatist,
is academic in nature. In this proposal for example, we have included
a biblfography of 89 papers in the field, organized into appropriate
sets of references to accompany each section. This obviously is not
an exhaustive bibliography but is inCluded as an approach to the
literature and existing techniques.
The end results of the tasks of the proposed contract will be
a comprehensive document on data processing technology and systems
aspects, a meaningful
and useful document to the system planner.
Each research area or new technology wi 11 be discussed from the point
of view of its meaning to efficient and economical computer systems
or its application In anti-submarine warfare, intelligence. electronic
countermeasures or whatever the functional requirement. It wi 11
explain what is happening in the field and wi II develop. analyze and
evaluate the various approaches. Finally, the processes will be
documented comprehensively and comprehensibly.
The project personnel must understand future Navy uses of the
data processing techniques and they must have imagination which leads
~
I
~
I
P9958-079
Page 1-3
them to judge what is likely to be important for the future.
This can
only be accomplished by unusucl1y highly qualified and experienced
data processing personnel.
thus qualified.
The Informatics Inc. team we be1 ieve is
Wu further believe that thfs proposal indicates our
great Interest in this project and our capabi lities for performing
capably.
•
•
I
I
P9958-079
Page 2-1
2.
SUMMARY
This proposal by Informatics Inc. covers two alternatives:
Subtask 2
only (Current and Projected Technology), or Subtasks 2 and 3 (Current and
Projected Technology, Study Integration and Design Methodology.
Consulting
will be obtained from Hobbs Associates In the area of computer hardware.
Our over-al
J
approach to the task is one of a combination of the prag-
matist and the researcher.
We bel ieve that the project team members must
understand the practical future operational uses of equipments but, on the
other hand, must also take an academic approach to investigating and
developing new techniques.
It will be important, for example, to survey
the literature on the various technical areas.
In discussing the various
technologies in this proposal we have developed a preJ lminary bibl iography,
since at the end of each section there is a bibliography on the subject
matter discussed.
j
These are not comprehensive bibliographies but onJy
Ilustrate the kinds of reference material to be examined.
Section 3, Research Methodology and Technical Approach, presents the
over- all description of the technical tasks to be accomplished.
There is
first a discussion of the interfaces among the various subtasKS.
We have
discussed the importance of understanding the technical aspects of Subtask
I on requirements.
Project personnel must understand the character of the
functions to be accomplished in future tactical command and control systems
and the implication of these in the requirements for data processing, hardware and techniques for designing and using it.
Figure 3-1 summaries briefly
some of the characteristics and requirements for various tactical command
and control system functions.
Also, an example of man/machine coordination
is discussed to illustrate the interplay between the various subtasks of
the project.
Technology research in the data processing field will consist of
identifying technologies and techniques of
potenti~l
after, analyzing, evaluating, and documenting them.
appJ ication and, there-
We have discussed in
Section 3 the various tasks to be accomplished under the general subtask
~
I
~
I
P9958-079
Page 2-2
of research in current and projected technology.
We visuaJ ize a compre-
hensive reference document wherein the various technologies are described,
analyzed and evaluated, and where certain trade- offs between approaches
are discussed as well as the applicability of certain approaches, of the
recommendations and their future use·
In Section 3.4
~tudy
integration is discussed.
Three different exampJes
are given to Illustrate the kinds of trade-offs which will be important
in this task.
Also, a summary of certain equipment and design trade-offs is
presented in the table of Figure 3-7.
It provides a candidate Jist of
technologies where design trade-effs must be thoroughly analyzed and understood.
A discussion of system design methodology is presented in Section 3.5.
Figure 3-8 presents a general implenrentation procedure for on-line systems
which will be used in future command and control systems.
We believe it
illustrates our appreciation of the over-all task of developing a concept
and of accomplishing the other major tasks of system design, system specification, programming and testing and modification.
Additionally, two examples are presented on approaches to the develop·
ment of a system design methodology.
Sec t j ons 4 th rough 6 cove r six maj 0 r tec.hn i ca 1 a reas of da ta hand I i n9
systems:
computer organIzation, Input/output and displays, hardware tech-
niques, memory techniques, programming, and advanced usage techniques.
We
bel jeve they represent an exhaustive list of technologies to be considered.
In these sections hardware and software considerations are considered together.
We believe this is consistent with the modern viewpoint on date
handling systems:
and software.
the Interchangeability and trade-offs between hardware
One of the major problems in all of the systems wi II be to
decide whether the programmer should be relieved of his burden by having
hardware to accomplish certain tasks, or whether flexibi lity and economy
is best obtained through implementation by software.
In Section 8 however,.
I
I
•
----_ ..
_--_ _ __ _._----------------------'
...
•.
..
P9958-079
Page 2-3
we have discussed programming from the standpoint of certain techniques
not directly related to hardware.
In the section on computer organization we have emphasized the trends
toward multi-computers and multi-processing.
We have not neglected, how-
ever, the new types of machines referred to as "highly parallel" machines
where the arithmetic and control logic is highly decentraJI,ed.
Also,
micro·programmed computers are discussed and it is pointed out that successful designs have emerged from a compromised micro·programming approach.
We bel ieve that analog/digital hybrid techniques could playa role, but
this
mu~t
be more fully investigated.
In Section 5 there is a lengthy discussion of input/output and
displays.
Displays and man/machine conmunication devices are becoming
extremely important as increasingly more functions of cQmJand and control
systems become automated.
of existing consoles.
Figure 5-3 presents many of the key characteristics
This is typical of the work which must be done in
understanding current and projected technology.
Likewise, in Figure 5-4,
a correlation is made between the application or functions to be performed
in a typical command and control environment and the hardware requirements
which result from those functions.
There Is also a discussion of the
various system configurations of attaching consoles to computer systems.
Group dispJays and communications are discussed tn Section 5.
We
point out for example, that the state of the art in group displays Is
lacking in many respects.
Customers are not happy with present equipments.
There are some uncertainties about the requirements and response times,
and there are many interesting technologies such as electro-luminescence
which look interesting for the future.
This will be a challenging and
contentious area for the project personnel.
In Section 5.6 there is discussed the very important area of communications.
Today's command and control systems and those of the future are
to a very great extent communications handling systems.
In this section
we discuss the functions, requirements, systems and uses in relation to
•I
•I
P9958-079
Page 2-4
communications and communications techniques and equipments.
In the section on input/output and displays, we have not neglected the
important area of computer programming.
In general the equipments are
general purpose and a considerable design problem is present in implementing these equipments through computer programming-
Programming systems
will emerge, for example, for display consoles just as they have emerged
for scientific problem solving.
In Section 6, hardware techniques are discussed from two basic
points:
stanG~
basic components and techniques and integrdted ci rcui tsmd vtnt!r
batch-fabrication techniques.
We have noted, for example,
Uie
research dnd
development effort in cryogenic components, magnetic Jogic components,
tunnel diode circui ts, kilomegacycle circuitry, and other more novel techniques of recent years.
Radically new techniques involving the use of
lasers and optical components are likewise discussed briefly_
We believe that integrated circuits and associated packaging
lechnique~
are not speculative future developments but are here today and should be
carefully considered for future operational use.
logistics, maintainability
and serviceabil i ty will be greatly improved by the use of integrated
circuits.
Memory techn i ques, as discussed in Sec t i on 7, are mos t i mpor tan t to
date handling systems since much of the technology in hardware and software
centers around the computer memories.
part in computer design.
Memory hierarchies play an increasing
Also, there are many techniques becoming important
such as the overlapped access of independent memories-
Addressing and
searching raises important questions about the use of associative memory
concepts.
There is no doubt that content searching, list processing and
the 1 ike will become important concepts in future computers.
We point out
that a number of important papers on associative memories and 1 ist processing devices recently appeared in the
)j
terature.
likewise the hardware
techniques of thin fi 1m and cryogenics are not to be ignored.
•I
I•
P9958-079
Page 2-5
Thin film memories are improving in cost and reliability.
Under this
section on memory techniques we have included a comprehensive discussion
of mass memory.
Much of this is summarized in Ffgures 7-1 and 7-2.
Recent) developments in mass memory techniques make them important elements in future command and control systems.
A very great cost element in command and control systems in represented
by prograrrming.
economizing on
Assemblers and compi lers represent the p05sibi II ties for
sy~tem
implementation.
However, selection of languages
and implementation of advanced compilers has not progressed
very
far at
this time, especially in areas of importance to real time programming.
The executive and master control concept for real-time 5ystems Is, however,
a subject of increasing importance.
detai I In Section 8.2.
I ikewise important.
This technology is
djs~ussed
in some
Understanding the total programming problem is
The experience gained on large mil itary command and
control systems such as SAGE, 465L, OPCON and the I ike should be carefully
considered.
Certain standards such as cost per instruction and the
"instruction yield per man month" should be developed.
We have included a short section on advanced usage techniques covering
areas such as ledrning and self-diagnosing techniques, heuristic problem
solving and language translation.
We point out that although these research
areas are, in general, not far enough along for immediate operational
consideration they wi 11, neverthless, have important influences on the design
of future hardware and software.
In Section 10 there is a short discussion of Naval tactical data systems
and National Emergency Command Post Afloat (NECPA).
A number of points
appropriate to this project are developed therein.
Section II, on the project plan, is a very important section since it
describes what work will be done, how it will be done, and by whom.
We
have included in Section II. I a detailed work statement and delivery schedule.
We have developed in Section 11.2 a work plan and milestones and SChedUles"
for the project as illustrated ;n Figure II-I. In Section II we point
I
•I
P9958-079
Page 2-6
out that a project control system to insure efficient and economic work
is of great importance.
It would be our plan to develop comprehensive
prQject plans to assure good project ffidnageMent.
The project team members proposed for this work are presented in
Section 11-4.
Dr. Walter F. Bauer is proposed Project Manager and
Mr. Werner l. Frdnk as Associate Project M,.mager.
p~uple
out the qual ifications of the ]}
be drawn.
As a group, they have had
control
system~.
exp~rience
personn~J
will
in many major command and
Experience in all lhe technical areas is
have surrrnarized some
and 12 - 2 .
from which project
Six of the 11 per~onnel propo~ed have had uver 12 year~ exper-
ience.
We
In Section 12 we point
of the
experierlce of the f.>er~onncJ
The p r f) j e c t per son nc} ha v cpu~.J I i ~ h(; d
applicability to this project.
0
re~re~t=nted.
in Fi~ure~
ve r 4 J pap t: r S
0
12-1
f d j rc c t
They arc li~tcd in Section 12.4.
Informatics Inc. and Hobbs As~ociates are both heavi Iy com~i tted
in mill tary data
izations.
system~
work for i
t
i~
the main orientation of
the
organ-
We are extremely interested in participating in this challenging
work fo r the Navy.
•I
I•
~-
-----------.--------.--- - - - -
P9958-079
Page 3-J
3.
RESEARCH METHODOLOGY AND
TECHNICAL APPROACH
In this section the approach to the tasks are described without
specific relation to the individual data processing technologies.
The
intent of this section is to show our appreciation of the total job,
our understanding of the relationships among the three subtasks, and
the efficiency with which we can perform the tasks as indicated by our
candidate approaches to the technical effort.
Our proposal covers subtasks 2 and 3, that Is. the technology
as well as the study integration and design methodology tasks.
If the
Navy does not wish to consider Informatics Inc. for subtask 3, Sections
3.3 and 3.4 which cover study integration and system design methodology
are less applicable.
However, these sections would still be useful in
illustrating the apprecTation of the Informatics team of the work tasks
of subtask 3, an appreciation fmportant to the qualifications to undertake subtask 2.
3.1
SUBTASK INTERFACES
The various parts of the study contract for tactical command and
control systems interact heavily.
To develop a list of candidate
techniques and technologies for subtask 2 the contractor must have
experience and an appreciation of the command and control operation and
have the capability of readi Iy absorbing the developments of subtask 1
(requ J rements) .
To illustrate some of the points of interaction between the
requirements subtask and subtasks 2 and 3 a chart is presented in
Figure 3-1 indicating some of the computer requirements and conditions
whfch are related to certain system functions.
In the small space of
the chart it is not possible to present the relationships in detail.
Consider, however, as an example, "Status of Forces and Resources".
This is essentially a fj 1e maintenance-fi Ie interrogation problem .
,
•I
•I _J
I
I
i
I
i
I
I
I
I
L_________.__ .____ ._.__________._ . . ___._. _____.__ . . ____. _.___.___ ._____._.________._____.__.___
!
CENTRAL COMPUTER
SYSTEM FUNCTIONS
AS IllTI ES
DIG ITAL
COMMUNICATIONS
DISPLAY
~ONSOlES
GROUP
01 SPLAYS
MASS MEMORY
REQUIREMENTS
Weapon System Control
Fast and arithmetic
Probab 1y 1i t t 1e
Moni tor i ng
Not likely
little
Navigation (navigational
sate1) i tes, inertial
nay 'gat ion}
Fast and arithmetic
Probab Iy 1 i tt Ie
Mont tori ng
Not Ii kely
little
Status of Forces and
Fi le and character
hand) ing ab iii ties
For COI'mland
reporting
Great
usefulness
Some 1 ike ly
Very great
Resources
Fi Ie and character
For eommand
Great
decisions
Great
usefulness
likely usage
hand 1 i ng
Su rve i 1 lance Sys tern
Operations
Character and
ari thmetic
Probably
Useful
Useful
Considerable
System Simulation
(training and test)
Fast, character,
and a r i thme tic
No t requ ired
for locally
generated inputs
Great
usefulness
Useful
Great
Planning
Fi Ie and character
most desired
Probably none
Great
usefulness
Some 1 i kely
Great
Strategy and Tactics
Figure 3-1
requ ired
Computer Requirements Related to System Factors
I
P9958-079
Page 3-3
I
I
3. 1
SUBTASK INTERFACE (Continued)
Information is received from lower echelons about the status of weapons,
materials, etc., over digital data links. The information
is filed in mass memory devices. Consoles are used to interrogate the
109isti(~s.
system in detai led w.ys. Group display~ are used to obtain gross or
total presentations, usually on a geographical basis.
There is much interrelation among all 3 subtasks of the total
project. In _ paper on Informatron Processing in Military Command,
Bauer ('~ recently presented the diagram In Figure 3-2 showing man-machine
coordination. This can be used as an example to il1u~trate the interplay
among the various subtasks.
In the first place. the functions to be
accomplished by man and machine are listed on the diagram. These in
turn can give rise to more specific technical functions such as data
acquisition, data retrieval and the like.
Information flow likewise
comes Into the picture since in this man/machIne coordination the data
volume, data rates, and frequency of operatIon are important considerations.
Concerning current and projected technology, the question Immediately
arises as to what kind of device wi 11 interface between the man and the
machine (computer).
How much information should be shown on the cathode
ray tube or display device, and what should be the siz$ and number of
characters displayed? Concerning study Integration, questions arise
concerning the techniques tor attaching the console to the computer.
Should it be attached to an input/output channel? Should it have its
own buffering system?
Should it have its separate computer to service
the console apart from the main computer?
Finally, of cours., there is
the question of developing a methodology for answering the questions
for developing quantitative measurements of the trade-offs between the
various possible system embodiments.
It is difficult to overemphasize the requirement that the various
personnel employed on the three subtasks should remain in close
communication at all times. The various measurement parameters develop.
*Numbers in parentheses refer to the bibliography at the end of
each sec t ion.
I
i
I
I
I'----
.
---------
.....--....-.-..
----
..
--------
----
I
I
•
._J
MAN/MACHINE COORDINATION
MAN
FIGURE 3-2
1----·
P9958-079
Page 3-5
II
!
SUBTASK INTERFACE (Continued)
3.1
in the technology studies are dependent on the requirements of the
integration subtask requirements.
It will be necessary for subtask 2
project personnel to determine the point of view of the system project
personnel.
For example, in the consideration of disc file memories,
the technology personnel wi 11 want to know the Importance of multi-access
fi les.
They wi 11 further want to know from systems oriented people how
many computers might be required to simultaneously access a fi Ie or how
many posltfoners should be moved simultaneously by the various computers
using the file.
3.2
TECHNOLOGY RESEARCH
Ideally, the completion of subtask 2 on current and projected
technology should result in documentation which can be used as a ready
reference for all questions of future system design.
planning capability.
It should update
It should cover all technologies, and list
till
the parameters and characteristics which the future system designer
would want.
It should also Include evaluations of techniques to allow
the system designer to make decisions concerning whether the techniques
and technologies are to be used and how they should be used.
More specifically, the following is a list of questions on
technology which should be answered by subtask 2:
1.
What techniques are avai lable to perform the various functions
and operations required of canmand anc control systems"
2.
How have these techniques been used in the past?
What degree
of success have they emjoyed?
3.
What is the likelihood of future developments along these lines!
Will the future see a growth of this type of technique or is a
new turn in the technology likely?
What physical limitations
such as speed of light, physical and topological layout of the
components, size, and weight, and the like will be the future
1 imi t i"9 factors?
•
I
I
•
L.. __ _
._------_._._-_._--_.•_---------'
P9958-079
Page 3-6
3.2
TECHNOLOGY RESEARCH (Continued)
4.
How can the technique or technology be used in future systems and In
future system or subsystem developments?
S.
What are the interfaces and trade-offs between this technology
and its counterpart In software (hardware)?
6.
What are the parameters which determine the applicability of
this device, and what are the various parameters and values of
the characteristics in present and projected technology?
Ffgure 3-3 shows the four work tasks in subtask 2.
On the basis
of inputs received from subtask I on data requirements and on a survey
of the technology, techniques and technologies are identified which are
candidates for consideration.
Lists of present techniques are developed
as well as techniques which appear to be necessary.
An evaluation is made
of the relative importance of the various techniques and the requirements
for analysIs are likewise developed.
With further reference to Figure 3-3. the next step requires an
analysis of the various techniques. The various techniques are studied
carefully on the basis of existing literature and verba! discussions.
As a result of this analysis the various characteristics are developed
and various application parameters are generated.
Technical discussions
and specifications are likewise developed as appropriate.
informatiCl"l on which to evaluate the techniques.
lhis provides
The evaluation work
task results In statements concerning the applicabIlity of the techniques,
recommendations for its future use and remarks on how the techniques can
be used in systems and in subsystems.
The figure a1so shows that the process is not a lInear one, that
the various work tasks, as they are performed, give
"previously" accomplished tasks.
proceed in
p~ral1el
to a
v~ry
In
great
ot"~r
e~tent.
rls~
to changes in
words, the tasks In reality
Also the various work taSKS
develop outputs Important to the study integrati0n phase of the work
I
I
even before the final documentation of subtask 2 Is completed.
I
I
L ____.
._---,--
•I
•I
.----.--------.. ------------ .--------------.-- - - - - - - - - - - - - - '
P9958 .. 079
Page 3-7
-.....
....
-
Data Requirements
fran Subtas k 1
Technology Survey
IDENTIFY
Delet e, add
Chang e
-
.......
Technology Lists
Importance Evaluation
Analysis Requirements
-
Existing Analyses
and Techniques
......
"
-
ANALYZE
"
.....
To Study
Integration
Task
-
Technical Discuss ions
Characteristics, Parameters
and Specifications
Evaluation Requirements
Modify Analysis
Requirements
..........
--
-
EVALUATE
To Study
Integration
Task
-
System Descriptions
Trade-off Analyses
App 1 i cab i 1 i t Y
Recommendations
U
-
~
DOCUMENT
" Comprehensive
Reference
Document
FIGURE 3-3
CURRENT AND PROJECTED TECHNOLOGY RESEARCH
I
L-_ _ _ _ _ __
--
P9958-079
Page 3-8
3.2
TECHNOLOGY RESEARCH (Continued)
All of the work tasks described here would apply to all of the
major technical areas of data processing.
They are as follows:
computer organization,
input/output disp1ay,
hardware techniques,
memory techniques,
programming, and
advanced usage techniques.
These technical areas wi '1 cover all of those whJch are candidates for
consideration In tactical command and control systems.
These are
discussed in Sections 4 through 9.
3.3
APPLICABILITY CRITERIA
3.3. I
Discussion of Criteria
A very important function of the proposed study will be the
development of criteria for us. by Navy planners in selecting the types
of components, techniques. and operations to use in each specific portion
of the 1970 - 1980 ear Navy Tactical Data System.
During the study,
whi Ie analyzing the different components and techniques that appear
promising for that time period, careful consideration wll) also be given
to determining criteria to use as a basis for comparing and evaluating
these components and techniques.
Criteria will be established for the
major components, techniques, and operations to Indicate the conditions
under which they should be used and their relative advantages and
i
disadvantages.
For example, charts plotting the cost of magnetic core
and thin/film memories as a function of capacity and speed can b. used
iI
I
I
I
I
I
j
as a guide in determining which type of memory to use in a specific
type of application.
In establishing such criteria and making decisions on the basis
of them, it will be necessary to consider not only the performance
offered by the component, technique. or operatIon but, also, to
I
I
I
I
II
L____
----------------------------------------------------
----
•I
I•
P9958-079
Page 3-9
3.3.1
DIs~usslon
of Criteria (Continued)
properly balance this
performan~e
against the cost, reliability,
maintainability, and environmental conditions. It will be necessary to
give adequate consideratIon to the ~ompatiblllty of each ~omponent and
technique with all others used In the system, and to give proper
consideration to compatibility with existing equipment, systems, and
software. Equipment and techniques developed and experience gained In
connection with the present Navy Tactical Data System must be adequately
weighed in proposing new components, techniques, software, or methods
of operation for future systems.
3.3.2
Research - Operational Lag Times
In addition to considering the advantages and disadvantages of
new and proposed components and techn,iques, and considering their
technical feasibility it will be necessary to accurately estimate the
date at which these new devices can be expected to be fully operational
on an economic basis. For example, It is not sufficient to determine
that associative memories can offer significant advantages In the operation
of a Navy Tactical Data Syst... It is also necessary to determine
correctly whether associative memories of appropriate size wi 11 be
avai lable on an economic basis at an early enough dste to pian their
inclusion in a Navy Tactical Data System for the 1970 - 1980 era.
Cryogenic techniques are an excellent example of this. As early .s 1956
research workers were proposing the use of cryogenic memories. Since
that time, consIderable effort has been spent in research and development
on memories of this type. Yet seven or eight years later, no cryogenic
memories are in commercial use,aJthough they sti 11 offer great promise.
This Illustrates the necessity for knowledgeable consideration of the
technical problems involved that might further delay the su~cessful use
of new techniques in an operational system.
•I
•I
P9958-079
Page 3-10
3.4
STUDY INTEGRATION
The study Integration phase of the work is probably besl
accomplished by Identifying certain candidate systems and lubsystems.
For .. ch of these different approaches to accomplishing the data
proceSSing tasks, the various trade-offs are Inve.tigated to develop
the proper combinations of systems effectivenes., cost, maintainability
and the 'Ike. Sometimes the approaches will consist of various ways
of approaching the over-all system, and In soma cases an Input/output
system might be considered. In still other systeMS various component
technology combinations might be considered.
are
8.
I.
Just a few examples of iome of the ways of approaching the task
follows:
In multt-computer approaches there are two basic ways of
handlIng the comMUnfcatfon wIthin the system. The two ways
sre shown In Figure 3-4.
The various trade-offs among reliability, cost, aval lab; Ii ty
and the lIke, can be I.ses.ed for each of the.e systems and
• comparison can be m.de in a clear-cut w~y which establlsh.s
the merits of one system over the other.
2.
There are two basIc W8YS of handlIng the Input into a
commend ~nd control system of communlc~tlon5. One of these
referred to as " J nterna 1 mu 1tip 1ex; 09" and the other 85
It ex te rna 1 mu 1t f p 1ex t ng" .
I n t e rna 1 mu 1tip 1ex I n9 i rN 01 v e S the
r~d-fn of communfcatlons InformatIon dIrectly Into the
'S
computer which Is to process it. External multlptexlng r.fers
to the receipt of the communications infonmatlon In computers
or buffers which are external to the computer processIng it.
To aid In the analysis of the two basic approaches we must
consider some fundamentals of queuing theory Involving such
variables as processing speed and the delay of .ervlclng of
'ncomlng messages. Figure 3-5 shows a re'atlonshlp between
•
•
I
I
TOPOLOGY OF COMMUNICATION
PROCESSOR
#1
PROCESSOR
#2
PROCESSOR
#3
SUBSTATION COMMUNICATION
PROCESSOR
PROCESSOR
PROCESSOR
SVv'ITCHING
PERIQUIP
PERIQUIP
PERIQUIP
PERIQUIP
PERIQUIP
CENTRALIZED COMMUNICATION
FIGURE 3-4
"QJ
to
~
BASIC WAYS OF HANDLING COMMUNICATION WITHIN THE SYSTEM
W
I
"'0
cD
cD
CI'I
00
I
~.
cD
-------"---------"------"-----------,
P9958-079
Page 3-12
3.4
STUDY INTEGRATION (Continued)
various factors In servicing incoming messages. Given a
processing speed, a demand rate, and a unit processing time,
the average delay for message processing can be determined
from the graph.
This topic was briefly discussed 'n a paper
by Bauer (2), on computer design.
3.
As a third example, consider the attachment of console displays
to a computer system.
There are three basic ways of accomplishing
this as shown In Figure 3-6. In one technique, the console Is
tied directly to the Input/output device of the computer. In
another,as shown in the figure, buffering equlpments are used.
In a third technique a computer is used instead to pratllde
buffering. Having determined these three candidate approaches,
the benefits and applicabi Iity can be assessed for each in
relatIon to the others.
A summary of certain equipment and design trade-offs Is presented in
the Table of Figure 3-7. In general the left-hand column represents a
trend which is generally considered desirable In modern systems. However,
always there are advantages and disadvantages when comparisons are made
wi th an oppos i n9 or a J ternat ive approach.. I n the study I ntegrat ion
work, comparisons such as these will be made in great detai 1 with appropriate
analyses which yield quantitative results.
3.5
SYSTEM DESIGN METHODOLOGY
Figure 3-8 shows a general implementation procedure for on-line
systems such as those involved in command and control. The figure shows
certain management functions as well as the documentation required in
the orderly implementation of such systems. Descriptions of each of these
tasks have been prepared by Informatics Inc., but for the sake of brevity
they have not be.n included with this proposal. The figure does not
show a design methodology but rather shows the tasks which must be
~
I
~
I
3/2
I
I
I
I
I
I
AVERAGE DELAY
UNIT PROCESSING TI·ME
I
1/2
I
I
I
I
I
2
3
PROCESSING SPEED
AVERAGE SERVICE DEMAND RATE
FIGURE 3-5
QUEUING THEORY APPLICATION
P9958~079
Page 3-14
c
a)
b)
c
D
. c .•• -~-~
:._-G)
,.-
B
~)
~-.o
·Oj
I
I
d)
c'.
1
.
c
•
. I
I.-
B
e)
•
B
f)
C-
COMPUTER
- - UNlUfFE"ED I/O CHANNEL
g:a
BUffER
----- 8UfFE"ED I/O CHAN~EL
o ~ ON-LINE DEVICE
FIGURE 3-6
COMPUTER/ON-LINE DEVICE COWF1GUAATlOHS
I
~--------------------------------~--------------------------------~
I
EQUtPMENT OR CHARACTERISTIC
DESIRED OR CONSIDERED
Parallel Processing
ADVANTAGES
--
----
----------------~-------------------------,
DISADVANTAGES
OPPOSING OR ALTERNATIVE
APPROACH
Higher speed for circuit
component speed
Higher costs for switching
and programming
Single-stream process·
Memory Hierarchies
Higher capability-cost ratio
Higher programming costs
Homogeneous memory
Crossbar type Switch
(for system communication)
Low cost and high transfer
Greater cost for small
number of modules
I nforma t i on bus
Greater flexibility,
multiple uses
Grea te r cos t
Single-purpose consoles
Multi-computers and
Decentralization
High system reliability
Possible programming ease
Switching requirements
Possible programming
complexi ty
Centralized computer
Input-Output Pre-processing
(independent of main frame)
Relieves main frame
Photographic Type Group Displays
Elaborate Display Consoles
(CRTBs, extensive keyboards,
rates for high number of
modules
ing
etc. )
Possible extra equipment
Multiplexing into main
de~i9n
frame
Grea te r f 1ex i b iIi ty
F as te r response
Electric type
Problem Oriented Languages
(for real time operations)
Lower programming costs
Lower system operating
efficiency
Assembly languages or
machine coding
Modul ar i ty
System growth and
maintainabi 1 i ty
Possible higher costs
SingJeness or uniqueness
Extensive, Elaborate, Interrupt
logic
Programming ease, system
efficiency
Cost; design problems
Simple interrupts or
none
-0-0
IlJ
U)
~1.O
ro
FIGURE 3-7
SUMMARY OF EQUIPMENT AND DESIGN TRADE-OFFS
-.•- .... _--_..._--_.- -
-
-_ .. _---------- -
..
-
---- -
._- ----------
(.oJ
(J1
co
I
10
-....,J
(J'I
1.O
--------------------------------------'
P9958-079
P_ge 3-16
MANAGEMEN1
-----~-----
CONCEPT
AND
ENVI RONMENT
-. ,- - - - - - - - - - - - -r
T--------
-ISYSTEM
I
I
I
I
I
I
DESIGN
I
I
SYSTEM
!
I
I
SPECIFICATION
:
/
O"RATlONA<
\
""
\\
1:/
V
i
I
I
I
I
"'",
/1/
~~~
~-~ _ L
" \
1
AND MODIFICATION
I
I
, / (PROGRAIolS
I
TESTING
I
I
I
I
I
I
PROGRAMING
I
~
J
SYSTEM
DESIGN
/
SYSTEM
ACCEPTANCE
/
TESTING
A
/
I
\
"~
I
I
I
I
I
I
I
SPECIFICATION)
I
I
(IMING AND
SYNCHRONIZATION
STUDIES
~~
i
)
-L _____ _
I
I
-+
I
I
I
-1
---~
SPECIAL:ZED
/
\
I
, ANGUAGE
\ SPECI-ICATIQIII
~..
___
~
J
INTEGRATION
OF PROGRAIolS
/'
_ _____ 1..
-'
:
DOCUMENTATION
AND
r RAI NI NG
FIGURE 3-8
GENERAl I HPLEMENTAT ION PROCEDURE FOR ON-L I HE SYSTEHS
P9958-079
Page 3-17
accomplished In implementIng a system.
Further, it shows the inter-
action of certain tasks and the order in which they are accomplrshed.
We believe it shows our appreciation of the entire problem of system
implementation, and that It structures the whole task in a way that
the various aspects (system design, system specification, etc.) can
be discussed intelligently.
procedure than really exists.
The chart shows a more simp1ified
Actually the major tasks do not run In
a para1 leI fashion; the results of systems design and specification
very frequently feed back to affect system design.
We visualize the work of System Design Methodology as the
development of a system and techniques for accomplishing the implementation of an on-1ine or real-time command and control system.
The
Methodology especially affects the parts labelled "System Design" and
IISystem Specificatiorlt; it further develops a comprehensive approach
to these sets of tasks.
The Study Integration discussed in Section 3.4
is aimed at bringfng together certain technologies for comparison and
trade-offs which are necessary for system design.
Modern computer systems for command and control can be
considered as processing transactions of various types which arrive
at various rates.
(every
50~100
TypIcal transactions are:
receipt of radar data
mi l1iseconds); processing a console request; receipt
and processing of command messages; and feedback data for weapon
aiming or control (every 10 ms. or less).
In developIng a design, the
computer system requirements must be fitted to the operational
requirements.
The following paragraphs present an approach to a design
methodology.
Figure 3-9 shows the total Information system design method.
The operatfonal requirements give rise to the identification of
certain data types, volumes of data, the processing required, and
the distribution of requests for service.
problem.
Together they define a
If an initial assumption is made as to the speed for
processing each of these types of data, the entire process can be
1....--_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ • _ _ _ _ _ _ _ _ _ . - -- .. -"'------.--
•I
•I
OPERATIONAL
REQUIREMENTS
--
-
DATA_TYPES
INITIAL
PROBLEM
SPEED
DEFINITION
---
---
DATA
,
-
VOLUMES
~
ANALYZE
_ AND
SIMULATE
-
PROCESSING
..--.....-.... SERVICE. DISTRIB.
---~
.
,
ASSUMPTION
--
CHANGE·
--
ASSUMPTION
SPEED
·ANALYZE
~ ___DE_L_A_Y_D_A_T_A_..r----...;....
DELAYS
AND QUEUES
SERVICE
INTOLERABLE
SERVICE
O.K.
SPEED
REQUIREMENTS-
EQUIPMENT
ANALYSIS
•
PROGRAMMING
REQUIREMENTS
L
EQUIPMENT
IMBALANCE
-=-
DETERMINE
EQUIPMENT
COMPLEMENT
TOTAL
---
SYSTEM
ANALYSIS
FIGURE 3-9
INFORMATION SYSTEM DESIGN METHODOLOGY
~
,1
DESIGN
COMPLETED
SERVICE
IMBALANCE
P9958-079
Page 3-19
analyzed.
Probably the technique used here is sumulation by large
scale digItal computer.
This gives rise to information on the
average length of delays which, when compared with the delay requirements
i~lpo5ed
deterrnin~
by the 0pl'rational characteristics of the system. will
whether the service is tolerable or not.
If the service is
intolerable, then assumptions must be changed as to the speed with
which the processing is performed.
Having detennlned that the service Is
tolerable~
and the speed
requirements, an analysis can be made to determine the complement of
equipment required.
After this is done, the total
from the standpoint of
progran~ing
~ystem
is analyzed
requirements to determine if the
design is acceptable in atl respects.
service, or in the amount of equipment,
Possible imbalances in the
c~n
result in a further
anolysis and a change in the equipment complement.
An analYSis was performed at Ramo-Wooldridge simi Jar to this
and was reported by Rothman (2); the results are shown in Figure 3-10.
The problem mix gives the
characteristi~s
of the preblenl.
For example,
50 percent of the problem required servicing requests which had to be
completed on an average of 0.5 minutes and with a total processing
time of 1.5 minutes.
Poisson distributions were assumed. and the
frequency of arrivals is given essentially by the abscissa in terms
of the average number of requests arriving translated to the number of
computers to handle this average load.
The zero percent curve shows
that if the number of computers equ[Jls the average load, then there
is a zero probabi 1ity of servicing the problem mix.
As an excess of
computers is applied, service improves as the results indicate.
The approach described in Figure 3-9 and described above is
an over-all approach which must be deve10ped In considerably more
detail to be usefu1.
There have been other approaches and tools
developed.
Simulation languages for computer programming have been
developed.
A simulation compiler for ease of simulating a new or
~
I
~
I
10
9
8
~~~~~~--~--~--~~~80% --~--~
V,O
.~O~
7 ~-..f---+--+--#---::I~K---#-'---:I~- PRO B A B J LIT y 0 F S E RV I C I N G A
NUMBER OF
COMPUTERS 6
REQUIRED
t----4~L..-.t-~L..--~_#__'7!II_-_#_-_+_-
FIXED PROBLEM MIX*VS LOAD
AND CO M PUT E RCA PAC tT Y
5
4
3
*PRO'BLEM MIX
50% LOAD.5 MIN.MEAN TIME 1.5 MIN. THRUPUT
40% LOAD 2 MIN. MEAN TIME 7 MIN.THRUPUT
10 % LOAD 5 MIN. MEAN TIME 20 MIN. THRUPUT
2
0
-0
2
3
4
5-
6
7
COMPUTATIONAL . LOAD
8
9
-i 0
"V-0
Ol .CD
c.c
C1»
CD
<1\
Q)
W
I
FIGURE 3-10
QUEUING ANALYSIS EXAMPLE
N
0
I-
0
~
cD
P9958-079
Page 3-21
nypothetical computer has been developed by T. Sanborn at Space
Technology L~boratorles. It is called SIMCOM. A computer program
has been prepared to simulate and analyze processing systems.
This
'S
known as the "Gordon Simulator" (3). These wi 11 be examined, as
well as others, and certaIn procedures wi 11 be developed and recommended
for use.
•I
•I
L-_____ _
---
. . ._____ . _ _ ._ _ _ _ _ _ ._ _ _ _ _ _ _ _ --1
P9958-079
Page 3-22
REFERENCES TO SECTION 3
1.
Information Processing in Military Command, W. F. Bauer,
PROCEEDINGS, ACH National Conference, Syracuse, September 1962.
2.
The RW-400 Data Processing System, S. Rothman, PROCEEDINGS,
Auto-Math Conference, Internationa1 Congress of Information
Processing, Parts, June 1959.
3.
A General Purpose Systems Simulation, G. Gordon, PROCEEDINGS.
Eastern Joint Computer Conference, Washington, D. C.,
Decembe r 196 1.
•
•
I
I
---------_._------
P9958-079
Page 4-)
4.
4.1
COMPUTER ORGANIZATION
INTRODUCTION AND DEFINITION
Computer organization is the newest and fastest growing area of
technology of computer design.
Whereas logic design refers to the design
and interrelation of components and circuits, computer organization refers
to the design and interrelation of larger system elements such as memory
units, control units and the like.
Computer organiLation emphasi.:.:es
the
point of view of the system designer and the user.
Computer organization will be an extremely important item in
tactical command and control systems.
I t is closely related to use of the
computer in the total command and control system.
The choice of the
computer organization wi 11 have a heavy influence on system responsiveness
and cost as well as on the total operationa1 characteristics of the command
and control system.
Some of the questions or considerations involved in computer
organization are as follows:
1.
How should data processing functions be divided among the
various computer units?
2.
What should be the degree of modularity of the system?
3.
To what extent should the system have parallel ism in the
operation of its major components?
4.
How is switching and communication effected within the
system?
5.
What should be the degree of decentralization of the system?
The above questions have a profound effect upon reliability, cost,
maintainability, and programming ease.
Since system reliability can be
achieved by dupJ icating certain modules, the choice of modules and the
•I
•I
_ _ _ _ _ _ _ _ _ _ _ _ _.__._ _ _ _ _ _ _ _ _ _ _-1
P9958-079
Page 4-2
4.1
INTRODUCTION AND DEFINITION
degree of modularity is important.
by having modules of similar types.
(continu~d)
Likewise, maintainability is improved
Cost is affected since, if a high
degree of modularity is achieved, reliability can be bought with the
addition of just a few redundant modules rather than duplication of the
entire system.
Also, costs are affected by the degree of central ization
since highly decentraliLed computer systems have "overhead costs" in switching
and communication equipments.
Last, but certainly not least) ease of
programming is very much dependent upon how the above questions are answered
since a certain amount of paralle1 ism in computing systems simp1 ifies
prograrrming whereas extens ive parallel ism comp1 ic.;ates it.
4.2
MULTI-COMPUTERS AND MULTI-PROCESSING
There is a considerable trend toward multi-computers and multi-
processing.
The term multi-computers refers to the attachment of a number
of computers or major components in an integrated system whi1e multiprocessing refers to the 'Isimultaneous " processing of problems through
multi-computer systems or through time sharing.
Mu1ti-processing allows
a higher duty cycle of the variuus major computing units.
It allows high
speed to be achieved by dupl ication of units without pressing circuit speeds
or approaching PQysical limitations such as the finite speed of light.
Multi-processing is often referred to as multi-computing.
Time sharing is another important technique.
Time sharing is not
as intimately involved in some of the over-all system considerations as
multi-processing obtained through multi-computers.
Time sharing will be
important as a programming technique to gain adequate response of the
computer system to communications devices, peripheral equipments, and
man/machine interface devices such as display and interrogation consoles.
Some of the country's more interesting time sharing projects are underway
at System Development Corporation and MIT.
•I
I•
P9958-079
Page 4-3
4.2
MULTI-COMPUTERS AND MULTI-PROCESSING (continued)
Multi-computers have been defined generally by Bauer
(J)~
In
that paper four main criteria are presented which delineate the degree of
parallel ism achieved among memory and arithmetic units before the system
can qualify as "multi-computer".
computer systems.
The paper describes types of multi-
For instance there is a choice between the hierarchy
and distributed concepts in the assignment of functions as illustrated by
Figure 4-).
Likewise, there are the substation and centralized approaches
to communication in considering the topology of communication.
Finally,
communications and switching techniques are divided into the information
bus and the centralized switch types.
advantages of multi-computers:
The paper goes on to explain the
lower equipment cost, reI iability,
expansibility and flexibil ity, control hierarchy establ ishment and ease
of programming organization.
Each of these advantages is discussed in
some detail.
Multi-computers are in use or are being planned for immediate
use in p resen t day Nava I tac t i ca 1 da ta sys terns. The comb i na t i on of USQ-20 IS *-I(
used in the Naval Tactical Data System as explained by Chapin (2), underscores
this use.
Also, there is extensive use being made of multi-computers in a
NavY,system being p1anned at the Pacific Missile Range as explained by
Bauer and Simmons (3).
In both of these uses part of the mUlti-computing
characteristic of the system is achieved through high speed digita1 data
links connecting remote computers.
Significantly, in both appl ications,
computers of the NTDS family are used with multi-computers at the nodes of
the networks and with the nodal points connected with the high speed digital
data 1 inks.
4.3
HIGHLY PARALLEL MACHINES
One of the newer aspects of computer organization can be referred
to as the "highly parallel machine" organization.
*
This organization
Numbers in parentheses refer to the references at the end of each
section.
** Official name CP642A/USQ-20
j
1-.-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . ______________________ ._ _ _ _.. _ _ _ _ _ .... ________.. _ _ _ _ _ __
•I
•I
FUNCTIONAL ASSIGNMENTS
PROCESSOR # 1
SPEC IAL
PROCESSOR
PROCESSOR # 2
SPEC IAL
PROCESSOR
#1
SPEC IAL
PROCESSOR
#3
#2
SPEC IAL
PROCESSOR
#4
HIERARCHY CONCEPT
r
-PROCESSOR-··~....- - - - - - - - - - - - t
#1
PROCESSOR
#2
PROCESSOR
PROCESSOR
#4
#3
DISTRIBUTED CONCEPT
""0
01
1.0
FIGURE 4-1
(I)
~
I
-1:10
~
U)
U)
CI1
co
I
0
.....
\0
.- . .-.-. . . .-.
·---·--···-·--··--···-··--·----~--l
I
P9958-079
Page 4-5
4.3
HIGHLY PARALLEL MACHINES (continued)
represents an extrapolation of the parallelism ideas discussed in Section 4.2.
In these machines a network of arithmetic and control uni ts up to 32 x 32
matrix size or higher, is visual
i~ed.
Two of these computers have been discussed in the literature.
One is the Solomon computer discussed by Slotnick, et al (4).
Another is
the Holland machine or a modification of it, discussed by Comfort (5).
These machines are conceptual designs and the computers have not been built
or dre in only the early stages of development.
One of the important aspects of the contract work proposed is to
evaluate these types of computer designs for future command and control
activi ties.
I t appears that these computer designs wi 11 not satisfy the
requirements of the central computer operations in these command and
control systems.
However, there is a distinct possibility that the
computer designs will prove efficient for certain special ized operations.
At the present time the designers of these highly parallel
machines point to the solution of equations of the diffusion type, such as
those involved in weather prediction and neutron reactor flux analysis.
However, there is a possibiJ ity that these computers can be used efficiently
for such operations as the correlation of radar information thus making
them good candidates for command and control system usage.
4.4
STORED lOG·1 C AND MI eRa-PROGRAMMED COMPUTERS
Completing the spectrum of the different kinds of computers that
are in use or are visualized are the stored logic and micro-programmed
computers.
Micro-programming has been a technique which has been discussed
over the last 6 - B years.
It consists of the design of a computer which
the programmer can program at a micro-instruction level; using this computer
he can construct his own
instructions.
macro-in~truclions
or conventional computer
The advantages claimed for this technique are the
•I
•I
I
__. _. __.___..... ____._._. .__._._ . . ___ ._.___. ______. . _. ._. ___ ..........__. ____ . ._.______. . __....._. . . _. ._._ ......___ . _._____.______. _. _._ . _._.___ . __ J
P9958-079
Page 4-6
4.4
STORED LOGIC AND MICRO-PROGRAMMED COMPUTERS (continued)
flexibility of application through the construction of an arbitrary
instruction set, the simplicity of the hardware design through the reduction
in the number of logical operations which need to be performed, and the
resulting lowering of
co~t
through these benefits of increased flexibil ity
and simp 1 i city.
One of the "purer" approaches to micro·progranmed computers was
explained in a recent article by Timofeev (6).
He stated that one of the
disadvantages of stored logic computers -- a slower operating computer than
a computer with equivalent circuit-component speed _. can be removed by
an optimizing technique.
The "pure" approach has not gained much favor.
However, recently designers have actually built machines along micro·
programmed I ines, but have compromised the purer approach by adding.
instructions of the macro-type such as those in conventional computers.
An example of this is the PB 400 described in a recent paper by Boutwell (7).
This computer uses two types of memory system in which one of the memories
services the micro-progranmed operation of the computer.
Significantly, a micro-programmed computer (or stored logic
computer) is used in Naval tactical data systems.
It is -che AN/UYK-l
des i gned and bu i 1 t by Ramo Woo I d ridge) 'and used ex tens i ve I yin the
navigational satel1 i te program.
I t uses micro-progralTlTled ins
~ructions
and macro-type inSLructions and is, therefore, not a pure micro-programming
approach to computer design.
Still another technique of interest in micro-programming is the
use of read-only memories to implement the micro-programmed logic.
To
cycle through the various small micro-programming steps in executing an
instruction, memory devices can be used which dictate the micro operation
choice and sequence.
Since there is no need to read into these memories
certain high speed storage techniques can be used for implementation.
The read-only stores frequently require one microsecond or less, using
•I
•I
P9958-079
Page 4-7
4.4
STORED LOGIC AND MICRO-PROGRAMMED COMPUTERS (continued)
standard ferrite memory techniques.
The design of such a computer was
described and evaluated by Wheeler (8).
ANALOG/DIGITAL HYBRID TECHNIQUES
4.5
It is not I ikely that pure analog techniques wi I I find their way
into Naval tactical command and control systems except for very special
purpose devices of weapon or instrumenl control type.
However, analogi
digital techniques represent a segment of the computer technology which
is
at least a candidate for use.
The advantage of an analog computer is that a great deal of
efficiency is gained in certain types of problems through highly parallel
operations.
In other words, many integrating amp) ifiers connected in a
circuit can operate simultaneously and achieve efficiency through this
pardJlel ism.
These systems are especially efficient on problems involving
ordinary differenlial equations such as those in trajectory analysis.
Since many systems also involve digi tal techniques, and since the digital
computer is better used for higher precision operations, data manipulation
operations, and decision processes, the combination of analog/digital
sometimes appears attractive.
In the 501ution of differential equations, such as
tho~e
in
trajectory computations, the analog computer can handle high frequency
variables in the integration
prucesse~
whereas the digital computer can
provide the high precision requirements.
The combination can be used in
interesting ways to achieve highly efficient differential equation solutions.
One of the country's earl iest analog/digital systems was described in a
paper by Bauer and West (9) in which a number of these considerat ions were
discussed.
One of lhe interesting aspects of the proposed project will be
to ascertain if and where analog/digi tal systems can be used efficiently
in Tactical Command and Control systems.
•I
•I
P9958-079
Page 4-8
REfERENCES TO SECTION 4
1.
IIWhy Multi-Computers?", W. F. Bauer, DATAMATION, September 1962
2.
"Organizing and Prograrrvning a Shipboard Real-Time Computer System",
G. G. Chapin, PROCEEDINGS fall Join~ Computer Conference, November 1963
3.
liThe PHR Real-Time Data H.ndl ing System", W. f. Bauer and Sheldon
Simmons, to be publ ished in DATAMATION, January 1964
4.
liThe Soloman Computer", O. L. Slotnick, W. C. Borck and R. C. McReynolds,
PROCEEDINGS fall Joint Computer Conference, December 1962
5.
"A Modified Holland Machine", W. T. Comfort, PROCEEDINGS Fall Joint
Computer Conference, November 1963
6.
I'Microprogram Control for Digi tal Computers/l, L. B. Emelyanov and
A. A. Timofeev, Symposium on Advanced Computer Organization, IFIPS-62
Conference, Munich, August 1962
7.
liThe Logical Organization of the PB 440 Microprogrammable Computer l l ,
E. O. Boutwell and E. A. Hoskinson, PROCEEDINGS Fall Joint Computer
Conference, November 1963
8.
"Read Only Stores for the Control of Computers", D. J. Wheeler,
Symposium on Advanced Computer OrganiLation, IFIPS-62 Conference,
Munich, August 1962
9.
I
System for General Purpose Analog-Digital Computation", W. F. Bauer
and G. P. West, ACH JOURNAL Vol. 4, pp 12-17, 1957
IIA
l_________.____._____.____.______ ._. _.
•I
•I
P9958-079
Page 5- I
5.
5. I
INPUT OUTPUT AND DISPLAYS
DEFINITIONS AND FUNCTIONS
Input/Output and Display is one of the more interesting aspects
of data processing for tactical command and control systems.
are obvious:
The reasons
computer systems in command and control are on-line systems
which receive information from many sources and must present it to many
destinations.
This requires a full range of instrumentation and data
hand) ing devices attached to the computer.
In the modern on-) fne system,
data must be formatted from the external sources, multiplexed into the
memory of the computer or buffering device, then read in and serviced.
For output it must again be formatted, read out of the computer and multiplexed through the outputting channel to the receiving device.
The input/output devices and techniques fundamentally provide
an interface between the computer and a device (or person) external to
the computer.
For Instance, the computer receives information from sensors
such as radar and radar receivers.
and digitized.
This data must be properly quantized
It receives information from humans.
In the past this
information has come through switches, punched cards and punched paper tape.
r n the fu tu re i t
'V'J
i 11, with inc reas i ng frequency, come th rough comp Iex
on-line interrogation consoles.
The computer must interface with communications
equipments to provide capabi lity of receiving data from remote sensors, people
or other computers.
Also, the computer must, of course, interface with conventional
peripheral equipment such as printers, tapes and the like.
likewise it
must communicate with displays, not only console displays employing devices
such as cathode ray tubes but also group displays such as electric and
photographic types.
Another interface might be referred to as the
instrumentation type which would include devices like analog/digital conversion
equipments, digital data registers and simi lar devices.
All these interface
•I
•I
elements are used in many combinations.
I
I
L____________ .____._.________ ._
.----•........-- --. - - - - - - •.
._J
1--------- --- --- --- -------------- -- -----------
-----
-~.-.----------
I
P9958-079
I
I
I
I
I
Page 5-2
5.1
DEFINITIONS AND FUNCTIONS
(Contjnu~d)
A comp rehens i ve in t roduc t i on to the p rob 1t;ffi of i npu tl ou tpu t
has been provided In a paper by Bauer presented in the IFIP 62 conference
The following
ill Munich.
para9raph~
briefly ~ur.lfI1arize what wa~ ~aid there:
Buffering control and synchronization.
This is the process of
holding computer words temporarily and of providing Ilreadyl and I'lock outll
signals which allow the synchronization of the computer with the device.
A significant trend is the increased use of the interrupt
Interleaved memory
acces~.
~ignal.
As a part of the buffer process,
cycles of the computer memory used for the input/output process are
interleaved with memory cycles of the main computational stream to allow
the automatic multiplexing of input/output data into the memory.
Assignment flexibility.
This refers to the abi lity to use
input/output devices In many combinations so that, for example, a tape unit
is not dedicated to a particular input/output channel but is switched to
the using device of the computer system.
Simultaneous input/output and computing.
This refers to the
technique of performing extensive input/output operations wni le computing
takes place.
This is an extension of the buffering and interleaving
memory cycle techniques.
A further step is taken in this technique
by
performing predigestion of data and parallel processing of input/output with
central computing.
Often it appears desirable to perform many complex
input/output processes and perform the processing of the input and output
data independently from the main computer.
Thus the main computer receives
only totally digested data.
Man/Machine communications.
A most rapidly growing area is the
use of a display interrogation console at which a person sits and communicates
with the computer in a manner convenient to him.
This allows the development
of a close man:machine relationship in which the computer becomes a close
intellectual partner of the man.
Ii
i
I
1
I
L_.____
•I
•I
P9958-079
Page 5-3
5.1
DEFINITIONS AND FUNCTIONS (Continued)
These are the areas of growth in input/output techniques.
These
are areas to be carefully defined, examined, and documented in the proposed
program to insure that future command and control systems employ these
techniques to the greatest practical extent.
The following sections discuss
these areas from system, equipment, and usage points of view.
5.2
INPUT/OUTPUT BUFFERING AND INTERRUPT HANDLING
If there is an area of system design that concerns the planner
more than any other it is the general problem of input/output.
It is
relatively simple to develop from known requirements numbers reflecting
required storage capacities, required access times to data, and required
processing times for a variety of operations as a function of machine cycles
or arithmetic steps.
It is another problem, however, to analyze and trace
the data interactions and to comprehend the operational effect of the various
design alternatives associated with input/output handling.
In particular,
this relates to the techniques for interfacing computer peripherals and
communication lines with data processors, which ultimately depends upon the
buffering process and interrupt features inherent in the equipment.
5.2.1
-' npu t/Ou tpu t Buffer.!Jlg
An important aspect of this project is the identification of the
various equipment parameters and features associated with I/O
channels and
1/0 handl fng, and the subsequent relating of these to the parameters
characterizing data on the one hand and to the data carrier on the other.
Examples of terms that are typically used in describing I/O
capabi lity are tabulated below.
Data channel, simplex and duplex
Device Controller or Adapter
Party Jines
•I
•I
I
I
_J
---..---..-.-.. -.--.-------··---·--·--·-------·----------l
P9958-079
Page 5-4
5.2.1
I npu t:'Ou tpu t Buffe r i n9 (Con t i nued)
Assembly and disassembly registers
Computer to computer transmission
Buffered transmission, partially and fully
Cy c 1est ea 1i n9
Memory partition
Number of devices per channel
Independence of operations
Simultaneity of operations
High speed data rates
low speed data rates
Automatic data collection
Register scanning
Multiplexer
Formatter, assembler, decommutator
Compa t i b iIi ty
Parallel character/word transfer
Interlaced operations
Saturation rates.
It is important first to find a basic set of such descriptors
which will serve the system deSigner both as tools for evaluation
alternatives and as standards for making equipment comparisons.
It is next
essential to review the spectrum of buffering configurations which range
from computer l/O channels, through special purpose multiplexing buffers,
to buffers that are computers in their own right.
that may be posed concerning the last point is:
An example of a question
what are the economics of
a sate) lite computer acting in a communication/buffer role?
In fact. when
should a satellite computer be of a different size or class, from the
central computer?
In this analysis it is important to indicate the consequences of
•I
•I
1----'---"--.---P9958-079
I!
Page 5-5
5.2.1
'npu t/Ou tpu t Buffe r!129. (Con t i nued)
increasing capability.
For example, it must be made evident that increasing
numbers of data channels and increasing data flow rates will ultimately affect
internal processing throughput (if for no other reason than an increase of
core cycle stealing).
The
most important contribution, however, will be to associate
the I/O interface attributes with typical Information that wi II be known to
the system designers in each application concerning the data and the
peripheral;;.
Thus preferred buffering schemes wi I I be advanced for such
examp les as:
hundreds of
m~ssage
originating stations,
video transmission,
block transfer of data,
volati Ie display refreshing,
samp led readou t,
continuous data collection,
and will also be looked at from the point of view of the devices themselves
including discs, typewriters, conventional peripherals, communication lines,
analog sources, etc.
5.2.2
Interrupt Handling
In addition to the 1/0 buffering discussed above the system
designer wants to know the function and effect of the computer interrupt
capability, especially as it relates to the external interrupt.
In general
this feature signals:
(a)
End of an operation - as with the termination of a buffered tape
transfer.
(b)
Beginning of an operation - as when it signals the presence of
data in a register, which may be overwritten with new data if
not recognized in time.
•I
I•
__________._ ._ .____. _. __.____.__._.___ . __________J
-- ----- --- ---
------- ---------
-- .. ------------------ - - - - - - - ,
P9958-079
Page 5-6
5.2.2
(c)
Interrupt Handling (Continued)
Status condition - as In the case of an alarm signal or making
a mark with a real-time clock.
These signals aid in the management of the housekeeping, timing and
scheduling inherent in on-line systems.
The increasing number of 1/0 channels, decentralization of the
processor to multi-computer and memory modules, and varying sources and
rates of input data have led to considerable sophistications of the Interrupt.
For example, In most modern computers the number of unique interrupt
locations has increased (to the thousands) and the concept of priority
sequencing of interrupts has been promoted.
The effect of this additional
capability is compared to the conventional processor in Figure 5-1 where
the hardware-software balances are shown.
The value of these modern interrupt systems must be presented in
the light of their utility to specific operations and system requirements.
In addition to citing the attributes commonly ascribed to interrupt
handling, an analysis wi 11 be performed to show how to use and assign
priority levels for typical problem mixes, involving digital and analog
readouts, real-time clock, I/O conventional interrupts, keyboard message
generation, etc.
Tables such as shown in Figure 5-2 will be valuable to
point out features and trends and to indicate the relevance of such
characteristics to system requirements.
The work on Subtask 2 and 3 will require a careful analysis of
interrupt techniques in view of their profound effect on computer and
programming efficiency. For Instance the Table of Figure 5-2 will be
updated and made more comprehensive.
An analysis will be made and
recommendations made for interrupt systems to be included In future
computers for use in Navy Tactical Data Systems.
I
I
I
•I
•I
L_______________________._________________________________________________________________________________
P9956-U/9
Page 5-7
COMVINTI-.AL
PTE ANAL
ON-LINE
MOe'S'•.
INTlUUPT
PAOCE SOA
IMPLICIT DISAlLI ••
IMPLICIT DISAILING
OF THIS AND LealEI --........
PR I OA I TV I NTIMUPTS
........- . 01 ALL I ma"uns
SAVI LINKAGE:
AUT
FIOM INTllt-
I
RU'TED TASK
- - - - - - - - -.- --
DETiMINE
AUTOKATIC
INTERRUPTING
CHANNEL" DEVIC
DETlMINI
CAUSE "f1F
INTEIlUPT
--~---~--
AELATE TO C~
RESPONDING ITEK
I N I/O ClU£U£
.
IMPLEMENTED
BY PROGRAMMING
DETEMIN£
ACTION'
I MPLEf'lNTID
IV 'ROGANV4tNG
- _.- - - - - --
l---------_T~~N·~UE
Of TEAM •• "1-
AUTOMATIC
IKPLEHENTEO,£AFOAM
•
BY PROGRAMMING
FUNCTION
... -.----------
~
£NAlL£S THIS
" LOII£R
" '''IGAITY
INTlARUPTS
EHAILE
INTERRUPT
EI.fAILE S Ai. L
{ INTERRUPTS
------------FIGURE 5-1 "
I INTERRUPT
HANGLING IN CONVENTIONAL" OM-LINE PlOCISS.S
Number of
Interrupt
Locat ions
Ccinputer
IBM
7090
Ext.
Int.
9
.-
3
-"""
Pri~ri ty
Sequencing
4 levels
-
Linkage
Oi sab Ii n9
Inhibition
Alert
Com-nen ts
All
Selective
i nd i v i dua I Iy
and
collectively
Yes
7909 prov ides 8 add i tiona I interrupt
locat i ens
Se lee t i ve
i nd i vi dua I 1y
and
collectively
fOt" I/O
Yes
II 0 p rio r i t y' Ieve 1s
Selectiv •
i net i v i dually
Ves
Collecti.vely
Ves
,
Automatic;
Stores IC
interrupts
only
UNIVAC
6S
9
8 leve Is
1107
CDC
7
2
.2· 'eve.ls
1604A
UNIVAC
33
1
8 "vels
1218
Prog rammed;
Use a return
jump
AII
interrupts
Automat i c;
Stores IC
only
interrupts
Prog ranned;
Use a return
jump
interrupts
All
All
in ch.nnel number
sequence
-
SIS
2-1024
UNIVAC
42
P.rogr~d~
levels
A II taller
Use a return
jump
pr i or i·ty
interrupts
5 level v
Progranmed;
Use a return
jump
A' 1
interrupts
2-1024
2
490
GE
M236
3-256
.3-256
I~els
Prog rammed;
Use a
special
i ns true t i on
except
fault
All lotler
priority
interrupts
Status word prov j ded
I 0
levels in channel
number sequence
wi th interrupt.
_.
920
I priori ty interrupt
CoIl ec t i ve 1y
No
Int.rru~t priorities
can be changed by
hardware change
Se 'ect i ve
individually
for 1/0
Ves
I/O pr i or i ty levels
in channel number
sequence
Selective
i nd i v i dua , Iy
and
col' ec t i ve 1y
No
Interrupt priorities
can be changed by
hardware change or
by . program
.,
""0
c.Q
(t
•
(I)
.--
INTERRUPT CHARACTERISTICS FOR A SELECTED HISTORY OF COMPUTERS
..
«D
Ut
Q)
(II
FIGURE 5-2
-v .
co·
,
0'
..
cD
~
P9958-079
Page 5-9
5.3
INPUT/OUTPUT DEVICES
The major types of convention al Input/output devices are:
punched cards, punched paper tape, magnetic tape, keyboards, printers,
and analog-to-digttal and dlgi tal-to-analog converters.
Any or all of
these may be applicable to the NTDS for lhe 1970-80 era.
However,
it Is anticipated that magnetic tape uni ts, keyboards, and printers
will be the preferred devices for deating wi th the human and digital
i npu t/ ou tpu t requ i reme n ts .
Anulog-to-digltal and digital-to-anaiog converters wi 1)
be used in dealing wi th other equipments that provide Inputs or accept
outputs In the form of analog voltages, shaft positions, or pulse
trains.
Punched cards or punched paper tape may be used for specialized
purposes, but for shipborne use they will probably not be as desirable
as Lhe other devices 1 iSled.
Keyboards and printers can now be operated on-1lne under
direct control of the computer thus e1iminating the need for punched card
or punched paper tape as intermediate storage for input/output functions.
Extensive mi 1 itarized peripheral equipment does not exist at
present.
However, attention is turning to this area.
contract for a militarized disc.
IBM has an R&D
Data Products Corporation hus a
sImilar contract to develop a militarized printer,! and tape units are
available as noted below.
Data can be entered into computers by properly buffered and
multiplexed on-line keyboards and printed out by on-1 ine printers.
Historical data and files can be wri tten on magnetic tape (or in mass
memories discussed in a later Section) to be read later by the computer
for further use.
Magnetic tape will also be useful for storing and
inputting programs, and for keeping a time-sequentiul chronological
file of all input/output transactions.
Militarized magnetic tape units
•
•
I
I
.-. -.- . . - -......-- . -.----..-..-.-.. . -.-...- .-. -..-.----.---.-....--.. .- '.--.-.-.-.--- - ..... --'--"--' ---·-·------·----··-1
1'- - ."- ."--- --'--" . - -.-. ..... _. -.- _. _- ..--_.
I
II
P99Sd-079
I
I
Paye 5-)0
INPUT/OUTPUT DEVICES (Continued)
5.3
are now dvai lable as the resul
progrdm
I
des~ribed
t
I
of a joint Navy, Army, and Industry
recently by Tyrrell, Morrison, and Staller (I).
During the study these conventional types of input/output
devices will be reviewed to
ascert~in
their usefulness In the NTOS.
The trends for future iMprovements in these devices will be analyzed to
provide a basis for estimating the performance that can be expected in
the time era under consideration.
Less conven ional input/output devices include display devices,
on-line digi tal data transmission links, character sensing dev;ces,
and paltern scanning techniques.
Display devices are
detai I in Sections 5.4 and 5.5 of this proposal.
di~cussed
in
The present NTDS
system makes extensive use of displays and digital data links (2).
Digital data transmission is covered in Section 5.6.
Extensive development efforts have been devo'ed to ch(]racter
sensing devices in recent years (3, 4).
com~ercial
on checks.
An example of a special ized
application is the use by many banks of magnetic ink characters
It is quite p05sible that furt.her inprovor.lents in more
gcner.)1 ized characLcr sensing devices will provide equipment useful
for input purposes in future
N~vdl
Tactical
D~ta
Sys:cms.
Developments in pattern or image scanning and recognition
techniques may be more important to future command and control systems.
This type of input coupled wi th internal computer pattern recogni tion
programs may permit the automatic interpretation of aerial phocography
and other types of tactical imagery.
One study in this area and some
of the problems involved hdve been described by Holmes, Leland, and
Richmond (5).
The Graphden device in the Army Tactical Operation Center is
an example of a unit that permits automatic input of graphical and
hand-drawn data (6).
I ...
The TACDEN device in the same system is an
•I
•I
.- .......
I'
"'-'"
..
'.'
..• _.
-- ..- _.... _.. --_ ..•.- .. - ..
_ _.__._
.......
...
I
- _....
....
"'---',
I
I
!
P9'.353-079
Ph,HH..ImE:ri r
f'leSSd:,Je
elll.ry uf data} forrnat.Li:lg,
vi~u,.tl
compo~er
·.hi-,l permi
vcrific8Liofl,
lS
kcybvard
and eJi ,i'l'.).
•
•
I
I
--------------------P9958-079
Page 5-)2
5.4
CONSOLE DISPLAY AND INTERROGATION
5.4. 1
Requirements. Systems and Uses
The incorporation of individual visual display devices into a
computer system had its origins with the SAGE project and the NTDS. In
this application the primary requirement Is to present the console operator
with a tactical situation, having a geographical relationship. From this
information decisions are made by the operator and the computer working
together.
The emphasis is the readout display and requires the hardware
feature of line drawing to Indicate, for example, vehicle tracks.
Simi larly,
such Interest Is exhIbited with the FAA and NASA although with a somewhat
different purpose In mind.
At the opposite end of the spectrum of console applications is the
man/machine communication function typified by the query requirements of
Command and Control.
An example Is the formulation of questions of a data
base concerning the status of forces and resources.
An early contribution
In this area was the console development associated with the Air Force Il7L
program. Here there was as much emphasis given to the Input process as the
readout.
Current Interest in such devices pervade the systems being
developed by the NMCS, NORAD and 473L, all of which are developing operational
systems using display consoles for the basic task of human query and response.
Pertinent to th I sproJ ect wi 11 be the I dent i f I cat Ion of app I i cation
areas together with user techniques to suggest to system planners the
potential of such devices and their limitations.
It Is, for example, important
to provide guidelines concerning such items as:
a.)
Amount of training required to operate a console;
b.)
Type of personnel {officer or enlisted} to be assigned as operators;
c.)
d.)
Degree of programming familiarity required;
Degree of familiarity required concerning particular data base
content and
e.}
~tructure;
Possibility of remoting device from the processing center .
j
I
j
!---------------
---------._-----------------
•I
•I
.. ----+--------.~-- . -------------------
------
------.---- . -- ..-
P9958-079
Page 5- 13
5.4.1
~!9uirements,
Systems and Uses (Continued)
In particular, the application
~reas
would Include data editing,
data entry, data base query, system control, system maintenance, on-line
analysts, on-line monitoring. etc.
The
import~nt
result of this study
will be the tdentification of console features necessary for the execution
of the cited applications.
Such consoles may range from a typewriter
keyboard to hIghly complex instruments such as the integrated console for
the roc of 473l.
5.4.2
Eguipments
As a result of the large amount of current interest in display
consoles several dozen manufacturers are now offering a variety of devices
for this purpose.
These range from strictly CRT or TV readout monitors to
consoles which fnclude background projection on the CRT (even with color).
full typewriter keyboards, light pens and, as in the case of TRW, ITT and
IBM, on·line modifiable labeling of special function keys (the so-called
overlay concept) .
Aside from such basic features the equipment environment also
involves detail features such as shown in Figure 5-3
for a number of
commercially available consoles.
A
ch~rt
such as this can however enly be useful if the physical
characteristics are related back to requirements.
to develop charts such as Figure 5 -4
Hence It Is necessary
, where the appl leabi Ii ty of
hardware features is shown for the indicated requirements of a specific
command and control system, the NMCS.
Associated with these equipment considerations Is the Important
question of console/computer Interface.
Here it will be necessary to indicate
problems and their solution involving:
I
i
I
a.}
Desirability of standardizing logic of interface, e.g., IBM 7291V
tape interface;
b. )
Effect of remotlng consoles, e.g., cables, amplifiers. etc.;
•I
•
I
L. _ .__.___________ - - - . ---.----. --.. - -.-.- - - -.- - -. -.-----..-.- .-. -----.- ".- .- -. .-.---.------------..---..-.- .- -.-----.--.-----~
P9958-079
Page 5-14
,
TRW
TRW
DC-400A
85
General
Infonaatlon
Data Display Inc.
Dyn~ics
SC-1G90
Electrada
DatacCD 408-2
00-13
Products Corp.
15041360314000
ITT
In·---A·.ad
~la
KEYBOARD CHARACTERISTICS
Control Keys
Character Keyboard
Func t i on Keys
45, Program Control 20, Fixed Control*
Fu J I Typewr iter
Full Typewriter
One Removable
Two Removable
Overlay of 30 Keys Over I ays of 30
Keys each
One standard, one
24, Fixed Control
8
46, Fixed Control
2, Fixed Control
None
None
Fu 11 Typewr iter
Full Typewr iter
None
None
None
4
none
1
Fu 11 Typewr iter
62, Switchable sets
of 30 keys
Opt ion
ENVIRONMENTAL INDICATORS
Status Indicators
24
30
Aud i b Ie AI arms
None
Yes
No
Yes
No
None
yes
no
Yes
Yes
No
None
4" X 8"
B1inking Signals
Intensity/Defocusing
CRT CHARACTERISTICS
Size of Scopes
10"
12" X 16"
Igtl
1911
17'1
Number of Mesh Points
20 X 36
384 X 512
256 X 256
1024 X 1024
16 X 63
(23 11 tube)
10 X 50
Two 19" Scopes
or
7 X 72
Repeti tion Rate
Program Control
30
or
Program Control
60 cps
Variable
60 cps
45 cps
max. 30 cps
Resolution Power
500 TV 1 i nes
Brightness
Reg is trat ion
50 ft.
Background Color
1- (i n Lab)
± 24% Char.
± 12% Char.
He i gh t
Height
Spot Size
4000 TV 1 i nes
20 ft. I .
± 20%
SiZG of Characters
±
Char.
He Igh t
S% Char.
Height
\
20 mi Is
2S mil 5
Light
Dark
Dark
62
62
64
CHARACTER REPfBTOIRE
Number of Characters
I 000 TV 1 I nes
;
20 mi Is
20 mils
Dark
Dark
light
Dark
46 Wi red
64
64
64
,
V"\
.3
Variable -
,
.100" + • 01 5"
18 ProgrMlDed
Variable
• 125 to .375
*
Max. No. of Characters
Displayed
20 x 36 - 720
32 x 64 - 2048
1000
Variation on Size
None
2 leve 1s
V.,*
Opt ions
16 x 63
4 Levels
FIGURE 5-, 'P'lrt I)
CANDIDATE CONSOLE tMARACTElISTICS
No
III:
1008
10 x 50 or
7 x 72
No
P995a-Q79
Pege 5...j5
I
TRW
TRW
DC-400A
85
General Dynamics
SC-JogO
'Data Display 'ne.
00-13
Info ....ation
Products Corp-
Electrada
OatacOil 408-2
ITT
Integrated Console
I 504/3603/4000
CRT USER FEATURES
Programmed
Progr . . .d
Hardware
No
No
No
No
No
M.A.
Light Gun
Hairline Cursor
Available
No
Ves
Ava i 1ab Ie
AvailabJe
Ves
No
No
V'\
2"
Full
M.A.
M.A.
No
Ves*
No
Yes
No
No
No
Ves*
No
No
No
No
Associated Buffer
No
4096 Words, 9 8i ts
Avai lab Ie
2048 x 24
Drwa or Core
3603 can Service
12 1504 1 5
Ves
Associated Hard Copy
Localized Maintenance
No
Yes
Avai 1able
No
Yes
No
Ava I Jab Ie
Yes
Ves
Yes*
Time Sharing Possibilities
for Multiple Consoles
Par i ty Check
No
No
No
No
Yes*
Yes
Ves
Character Generator No
up to 4 Consoles
Yes
No
Marker
Tab Function for Marker
Background Projection
LINE DRAWING
Maximum line length
311 If Arbitrary
DI rection
Yes
No
Color
SPECIAL DEVICES
CONSOLE OVER-ALL
CHMACIEB I SII CS
In 3603
Ves
Yes
CONSOLE PACKAGING
No. of Units
2
Size, Wxhxl
Console: 46
3
2
x 48lt
X
7211 x 51" X 33"
11
1504:
4411
Buffer: 46"
x 4811 x 24/1
3603:
3
14/1 x 14"
x 26"
4111 x 64'·
X 2411
4000:
Power Requirements
117 Vol ts AC
J 15 Vol ts AC
1 15 \101 ts AC
20S Vol ts AC
J 15 Vol ts AC
1504:
3606:
117 Vol ts At
117 Vol ts At
4000:
PRICE
Basic Unit Price
With Basic Options
Rent
*
Opt ions
$25K
$115K
$IS7K to $122K
No
$IS0K to $J43K
$48K
$23K to 8K
$42K
Ves
Yes
Yes
FIGURE 5-3 (Part 2)
CMDtDATE COUOlE tIUIllfEi·'.ITtCS
I
Approx. $20OK
OUTPUT
IF ON CRT
BASIC INPUT
~
IA
...
BEOU 18EMENTS
...eItJ
1
1\
-
11'1
11'1
VI
r:
o
.t-J
.j..J
:J
c:o
DAtA
8AS~
g
~
...,
~
V
~
l~
a..
«I
<..J
1\
0
0.1
t
0
(.J
'l:1
4.1
!l-
~
Q.
Q.
-a.>..
0
0')
11'1
'-
::J
<...>
(.!')
10
0
.:5'
~ _
"ID
~ IV
~·u
~
(lJ.....
0
14.. 14..
__
V)
Ol
C
Q.
L..
GI
--
0;;
11)._
e
L....e-!"-.I
L..
::t
.,..
-
Q
Q
~
...
II
~
~
Q,
"
_ .11'-1
~
tI)
~
11'1
-
.~
-0
c:
0
~
(.J
.2
4-J
4..1
ttl
Z
V)
~
g
._
L..
Q
IU
-
-
c:
U
Q::'
::s
~
~
"-
8a.
&
C
(IJ!3
0)_
-J;;c:
V).-
=:J
L..
-IoJ
0
()
11'1
~
IV
"_
2
~
x
:J
c::c
"0
(!J
g
v
oQ.
(lJ
11'1
__
-
'0
U
rn
0
V)
Cl)
10
M L..
tlJ
fl1
...c..c:
~
-
L.
u
0-
(l)
~....,
r:
c:
0
<..J
ttr
L.>...
EQJ
(I)....
e
c: _- . u2. .IlJ. , .o
!,:(!.l
o
..
0')
~
~
:
10"0
(I)
t:
0
...
0
U
tl1
."..
V)
~
-
0
-
c:
ItJ
_
....
~
-
U
'....
Q)
~
>(
0- 'Z
APP L I CAT IONS
~
0
~
.C
IA
~
ro
r:
0)
~
c5 :
~
~ 8
0)
r
::..::,"\
IV
0,
111
....
-
~
LI)....
o
V
C
'-
~
'-
0
C.J
~
fa
ID
~
::;
II
~
U
~
Q)
-1-'
'-'t1
~
U
SYSTEM
ASPECTS
L..
'""
~
...
OJ
11'1
II
~
CONFIGURATION
E
U
':1
L..
tlJ
~
'"0
t:t0
INTERROGATION
SOF/MOP
Weapon Totals
x
x
x
Uncovered Targe ts
x
x
Weapon Status Summaries
x
x
x
x
Weapon gel Ivery Analysis
x
x
x
x
Point Target Detai1
x
x
x
Fa) lout
x
x
x
x
Casua 1 t i as " F ac iIi t i as
x
x
x
Res i dua 1s
x
x
Oamage
x
Attack Pattern
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
BOA
x
x
x
x
x
x
DI Sf LAY AND ANALYS IS
Compo'Sltion of Ad Hoc Displays
Player Partleipatlon in War Gaming
x
Scheduled Outputs
x
Questions and Answers
DAIA
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
X
x
x
x
x
-.
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
INPUT
:
NUDET En try
x
x
x
x
x
x
x
x
St r I ket Launch En try
x
x
x
x
x
x
x
x
x
x
Exception Process ing
x
x
x
x
Damage Reports
x
x
x
x
Message Accounting
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
SYSTEM MONITOR. CONTROL
Queue Manipulation
x
x
x
x
Mod. So lee t i on
x
x
X
x
x
x
Equipment Env Ironment" 0 iagnostlcs
x
x
x
x
x
x
Startup
x
x
x
x
Breakpoints for Rollback
x
x
x
x
PROiAAt' DEBUGGING. MAINTENANCE
."
Q
Data Base Updating
x
x
x
x
x
x
x
dificatL.m
of specific functions be as much " professional' programMing
independent as possible.
Optimally. the system should be manageable
by the user. once the basic programming has been accomplished.
It is possible to separate the application oriented tasks from
those tasks that are general purpose and apply to all on-line
system applications and processes.
A division Is made between
the processes required in generating a message and the actual
procedures for executing the action that may be required. The
former concerns the mechanics of handling displays and composing
messages. The latter is concerned with actual fi Ie handling.
retrieval, processing. summarizing, and formatting. In this
discussion we restrict attention to the first aspect, the general
purpose processes.
L. ._ ._.______.___ ._.___. .__.__._. ______..____________ .____.____
•I
I•
P9958-079
Page 5-23
5.4.3
ProgrammIng Techniques (Continued)
The design objectives for the programming system are as fol lows:
1.
Provide general
appli~ations
2.
~apabillty
and flexibl Jlty so that virtual Jy all
can be accommodated.
Standardize technIques and procedures so that Individual program
segments or subroutines can be shared by as many functions as
possible.
3.
4.
Maintain order among contending users for the same files.
Service each station as if its operator were the only user
making demands on the
processor.
Based on tne above discussion the programming system must include:
Display Subsystem Executive Control
This program performs the basic scanning, sequencing, and queue
control for servicing the on-) jne devices.
In addition, it I inks
to the Master Executive Control which may be supervising the total
processing system.
The degree of complexity of the Executive
Control program wi 11 be greatly Influenced by the presence or
absence of such hardware features as a real-time clock and external
interrupts.
Func t i on Man I tor
This program maintains the history tables and establishes the
action sequences to be carried out as a function of the keys that
are depressed by the operator.
Utility Program Package
this is a collection of service routines used primarily by the
Function Monftor and Executfve Control.
The availability of these
general purpose programs precludes recoding of common functions.
User Language
Thfs fs the language which must be used by the application
programmer in writing his program.
The system must provide the
•I
;
-
I
I
----------~
1-----------------P9958-079
Page 5-24
5.4.3
Programming Techniques (Continued)
programmer with the ability to express his program In both the
symbolic language of the computer, where each
c~nd
generates
one machine Instruction, and In higher order languages, where
each command generates many machine Instructions.
To be effective. the higher order language must possess the
following chief attributes:
a.)
It must be powerful enough to express the application problem.
b.}
It must be such that non-programmers can use it with a minimum
amount of training.
c.)
It must be readily expandable so that new coovnands and functions
can be added.
The existence of a system as described above implies that
application programmers must conform to certain coding restrictions and
procedures so that all of the possible programs can be accommodated by
this approach.
Whi 1e this may seem a
dis~dvantage
It is, in fact, a saving
grace since:
a.)
't simp 1if i es
the prog ramm 1n9 because of the ex I s tence of sarv Ice
rout i nes.
b.)
It simplifies the implementation of a new application since the
desfgn must fit within the logical framework set forth by the
system.
The importance of the second point cannot be over-emphasized.
Without a well-defined organizational and procedural phi Josophy, the
programmfng design and Implementation of the individual application can
become much more complex and time consuming than is necessary.
Apart from the hardware and software requirements for support of
an
on~line
console system it is also necessary to indicate the load which
display consoles wilt place on total system perrormunce.
This Includes the
amount of core storage required on a continuing basis as a function of
I
I
L_________________________. _____c_. _______.__________----.-------------------------4------------.-------.. - - ________ _
•I
•I
------_ .._--------------------_ ...-_._-_.- - - - - - - -
------------
P9958-079
Page 5- 25
5.4.3
Programming Techniques (Continued)
number of consoles, the expected amount of transactions at each console
translated to required machine time over a period of time, and the potential
traffic problems that may occur with regard to 1/0 transfers.
With respect
to the latter point it wi 11 be desirable to prepare probability tables
reflecting waiting times for multiple console users for varying parameters.
An examp1e of such a table for the specific case of a ten console system 15
given in Figure 5-6.
Here is assumed a service time
of 1.5 and 0.9
seconds per request, with a thirty second interval between requests.
By using the theory associated with Poisson processes it Is
possible to estimate the waiting time from knowing the service factor.
This
number is the ratio of service time to total elapsed time between requests
and Is 1.5/30 or 0.05 for one of the cases.
The results of the traffic
analysis are given in Figure 5-6.
The results show that a waiting time will exist 19% of the time
for the 0.05 service factor.
Since the service time is 1.5 seconds the
average 'vlai t on the wai ting 1 fne wi 11 be
lightly under 3 seconds.
For
the second model a waitIng line wi 11 exist about 7.4% of the time.
The
5
average waft on the waiting line will be approximately one service time
of 1. 5 seconds.
Console Hardware TechniQyes
Hardware Techniques required in various types of console displays include buffer storage, symbol generation, cathode ray tubes,
photographic imagery, digital read-out devices (e.g. counter wheels,
NIXIE tubes, etc.), and hard copy printers.
Newer techniques under
development are photochromic dynamic displays (7) and electro luminescent displays (8).
In using these hardware techniques to design console
display equ~pments such as those described in Section 5.4.2, it is
necessary to determine the proper trade-offs between characteristics such
as capaci
tv,
speed, legibi I i ty, resolution, brightness, color, contras
l,
•
I
I
•
L__~__
r--------i
\
Number of
Consoles
0
Number of
Consoles
Wai ting
Probability of n consoles
r~ui ring serv ice
Serv ice
Serv ice
Factor
Factor
= .03
= .05
0
.538
.712
0
.269
.214
. 121
.057
• 121
.057
2
3
2
.049
.013
.170
.070
4
3
4
.073
.074
6
5
.001
.003
.00 1
.000
.187
5
.017
.004
7
6
1.000
1.000
.193
FIGURE 5-6
PROBABILITY OF N CONSOLES OF TEN
REQ.U I RI NG SERVICE AT THE SAKE TIME
!
i
L--
Cumu lat ive Wa it ing Probabilit}:
Service
Serv ice
Factor
Factor
= .03
= .05
• 192
.193
.014
P9958-079
Page 5-27
Console Hardwaro TechniQyes (Continued)
ratio, color capability, frame rate, and computer interface requirements
to determine the type of display that best meets the specific requirements of the problem.
In different systems some of these character-
istics will be more important than others.
The periodic regeneration
required in devices such as cathode ray tubes can be accomplished
by providing sufficient storage for an entire frame in the display
device or by periodically retransmitting the data from the computer.
In this case, a trade off is possible between the complexity and cost
of the display devices on the one hand and the required computer time
on the other hand.
For future Naval Tactical Command and Control Systems, it
will be necessary to evaluate conventional display techniques, special
cathode ray tubes such as the Printapix and the Charactron, and newer
techniques such as the light valve projector, photochromic displays,
and electro-luminescent displays.
Cathode ray tubes will probably continue to be the dominant
form of display for consoles although electro-luminescent techniques
may be feasible by the time this planned Tactical Data System is to be
operational.
The capabill ties and potential Improvements In cathode
ray tube displays will be analyzed to provide a basis for determining
the design criteria for console displays.
New cathode ray tube techniques,
such as the ability to project a photographic image on the face of the
tube to superimpose it over an electronic image, will be investigated.
The relative advantages and disadvantages of storage type tubes for
display versus standard cathode ray tubes wi th storage
f~nctions pro~
vid,d by another memory (e.g., a magnetic drum) will be determined.
Any anticipated changes in the relative meri ts of the different techniques by 1970 will be evaluated and recommendations will be made as
a guide for planning.
"
I
I
•
----- ---1
P9958-079
I
Page 5- 28
I
I
I
I
I
5.4.4
Console Hardware Techniques (Continued)
Keyboard techniques for interrogation will be studied wi th
particular emphasis on the question of whether to use completely
free manual keyboards or to use keyboards wi th a format control imposed
on the operator.
The methods of communications between the interro-
gation console and the computer, and lhe requirements for storage in the
interrogation console wi IJ have a direct effect on the type of keyboard
selected and the ease of using the console.
•I
I•
I
!
I
1__ •• _ _ _ _ _ _ _
___oj
r-·------·-· --.------.-----.----------- --- -- -
___________
•
_w
_
P9958-079
Page 5-29
5.5
GROUP DISPLAYS
5.5.1
1!~~~8nts andJ!~
Group displays range from static tote boards employing manual
updating procedures to dynamIc systems modified by computer signals.
These systems may have a response time varying from seconds If completely
electronic to several minutes if they employ electro-mechanical processes.
Photographic, electronic, or mechanical techniques may be used, or
combinations of them.
Typically, the requirements for such systems arise from the need
of making Information available on a continuing basis for monitorIng
purposes, as Is done In NORAD. or for purposes of simultaneous vIew by a
group of Individuals gathered together periodically as a decision making
body as. for example. the Joint Chiefs of Staff.
In this latter situation
formal briefings may be delivered or presentations given as a result of
ad hoc requests for Information.
The translation of the various requirements and operating situations
to display hardware will be an important result of this portion of the study.
5.5.2
Eguipments and Systems
To date military users have been dissatisfied and perhaps even
disenchanted with the group displays developed by industry.
In terms of
review there are perhaps three systems in operational use typifying stable
aspects of techniques. These are the Iconorama. Kelvin Hughes projection
system, and the Philco and IBM wall boards of the Defense Communication
Agency.
The first of these affords color through the use of multiple
projectors.
It Is electro-mechanical and has been known to be unreliable
and Inflexible.
The second represents a class of systems In which a
continuous roll of silver halold film Is exposed by a small CRT which is
under computer control, development takes place, and the frame is available
for view in 15 to 30 seconds. The third Is a static background with a
••I
iL-_______________.____ ._._____.
•I
_~
. _. ____ ..__ ._____. _ . .--. -.-..-.....-_... -.._.-.. _-.. - - - - - - - -
---
-
- - - - - - - - - - - - ----------------
P9958-079
Page 5- ~O
5.5.2
Equipments and Systems
fixed number of neons whose onAoff state and color selection is under the
control of the comput~r.
There is an interesting new fam. Jy of group displaYi emerging.
They are probably best characterized by the fact that they employ photo·
graphic techniques. The process results in fi Jm chips (unmounted single
frames of 70 mm fj 1m) which can be projected and viewed as $Iides.
principal characteriitics of these systems are as follows:
1.
They are on-line with a computer.
Other
The computer performs such
functions as the selection of background slides and the selection
and placement of data on the slide.
2.
3.
They are used for periodic bri ef i "95 of high ranking officers.
The processing consists of taking full color background slides
and placing overlay information in color on the backg round.
The number of symbols
tradition~11y
is 64.
Examples of this automatic slide generation type of equipment .re those
bui It by Ramo Wooldridge and International Telephone and Telegraph.
these samples the color separation process is used.
In both
Background pictures
can be taken with a color separation camera which provides three black and
white Images representing each of the three primary colors.
slide prepared by a co1or separation
proc~ss
To view a
white light is passed through
the three black and white images and then through appropriate filters to
reproduce the three primary colors.
Projection of the three colored images
to produce one image provides full color reproduction of the original.
In these systems overlay information is placed on the background
by a photographic process Involving the background fi Ims.
It is usually
desirable to blank out the overlay information so that there wi 11 be mixing
of the colors of
th~
overlay symbols with the background.
This is a complex
process but is accomplished, for example, in the Ramo Wooldridge system.
The Ramo Wooldridge systems uses Kalvar film which Is exposed by
ultra vIolet light. This fi 1m has excellent environmental characteristics.
It can be developed very easily by simply heating It after exposure.
The
three Images of the color separation process are then prepared with the
P9958-079
Page 5-31
5.5.2
Equlpments and Systems (Continued)
Kalvar fi 1m and become one film, probably of approximately 70 mm size.
The major difference between the ITT and RW systems is in the
method of exposure of the film to place the overlay information properly.
The RW system uses a mechanical technique to position the cnaracters on the
fi 1m.
The ITT system uses a Charactron tube to position the figures.
Since
the Charactron tube does not produce enough light for direct exposure of
the Kalvar film, an intermediate step of exposing si lver halo1d fj 1m must be
Because the RW system
taken before the final film product can be obtained.
uses mechanical positioning its response time per chip is of the order of
1-2 minutes.
The response time of the ITT equipment is on the order of
15 seconds.
It is significant to note. however, that in neither case of this
automatic film generation equIpment has there been a satisfactory system
developed from an operational point of view.
The RW machine Is installed
In the basement of the Pentagon and Is used by the National Military Command
System Support Center.
Although it works "satlsfactorily" it does not have
the reliabi lity needed for use by the Joint Chiefs of Staff.
The ITT
machine is reportedly in a similar or worse state.
There is a question about the future and value of these types of
film generating equipments.
When the ITT and RW contracts were let there
was a great deal of enthusiasm within the Department of D.fense for systems
of this kind.
This enthusiasm seems to have waned during the last year or
so, possibly because of the poor operational status of the equipment.
As a first step in developing an approach to displays it is
pertinent that users of systems described above (such as NORAD, SAC, NREC,
DCA, JCS) be consulted regarding the application and use of the equipment.
For example, it Is known that the projection system at DCA, analogous to the
Kelvin Hughes system described above. is not used.
On the other hand the
Iconorama at NORAD is the main presentation for CINCNORAO.
.--.------------ .------ "t- ------------------ - ---- --------- -
--
--- ----------- -- ------------ --------------------------------------------- --
-
•I
•I
---- ._._.. _----------------
P9958-079
Page 5-32
5.5.2
Eguipmen~~_and
Systems (Continued)
For systems such as described above it is intended to organize
8
state of the art sUmMary such as is suggested by the information in
Figure 5-7.
In addition to the indicated information in this table the
results of the review and evaluation wi I' be included.
Beyond this it will be most important to consider and evaluate the
work current1y under way as, for example, the light valve at GE and the
illuminated panels at RCA.
Hardware techniques are discussed in Section 5.5.4.
In addition to such equipment reports there will also be a set
of conclusions regarding operating characteristics and how they are
reflected in the hardware, includlng considerations of:
speed - the time lapse between the generating of data and its
proJ act ion,
clarity - the abIlity to register and produce saturated
symbols.
reliability - the development and use of
mean-time~to-failure
parameters t
cost - the total cost of implementation including programming
and offline procedures.
5.5.3
User and Programming Techniques
A significant aspect of making a disp1ay system operational is
the Joint effort of the end product user, the output designer and the
programmer.
Each affects the other and there is a continual feedback and
desire to modify, enhance, or make additions to the basic capabi lity.
In
such an environment it Is mandatory that a programming system be specified
which has general purpose attributes and is modular.
Hence, it Is absolutely
essential that systems of the kind represented by the NTDS have, as a basic
element, general purpose retrieval programs and output generators.
As an
example of a very desirable system with these attributes we cite the FLEET
Intelligence System developed at CfNCEUR and known by the name Formatted
•I
•I
.-.--.------.-.----------
I
J
P9958-079
Page 5-33
IlU
, • I " T
(I)
(2)
(3)
STORAGE AND RETRIEVAL
YES
YES
WITH MOOIF ICATION NO
COMPUTER TO SCREEN
TIME (in seconds)
10-12
30-240
II
COLOR
COLOR
LINE DRAWING CAPABILITY
YES
QUALITY ANALYSIS
VIEWING STATION
(4)
(5)
(6)
(7)
NO
NO
LIM I TED
REAL TIME
15
REAL TIME
I SEC/CHARACTER
APPROX.
COLOR
B" W
COLOR
B" W
COLOR
YES
YES
YES
YES
YES
YES
NO
YES
YES
NO
NO
NO
NO
PROJECTOR LOCATED
REMOTE FROM GENERATOR
YES
YES
NO
NO
NO
NO
NO
DELIVERED
YES
YES
NO
YES
NO
YES
YES
NUMBER OF SLIDES
REQUIRED FOR FULL COLOR
3 plus
1
2
NA
2
N A
Separate Projectors
for each Color
FILH
RECORD
I
background
NUMBER OF BACKGROUNDS
STORED IN PROJECTOR
100
200
300
0
-
NA
-
BACKGROUND ACCESS TIKE
(in seconds)
1
4
4
-
-
-
-
RANDOM ACCESS SELECTION
YES
YES
YES
N
BACKGROUND ONLY
-
BACKGROUND ONLY
SYMBOL COLORS
WHITE plus 3
BLACK, WHITE
plus 6
WHITE plus 5
BLACK, WHITE
WH~TE
BLACK, WHITE
WHITE plus variable
number (no black)
SYMBOL GENERATION TECHNIQUE
CRT
CRT
Silver Halide
OIL FILM
in Vacuum
CRT
Silver
KaJvar
DIRECT
CRT
Projection
SCRIBED
Silver Mirror
Ka lvar
DRY
Sten c i 1 Di ree t
to Kalvar
MASKING FOR UNAMBIGUOUS COLOR
NO
YES
MANUAL ADJUSTMENT NO
NO
NO
NO
RESOLUTION
GOOD
GOOD
VERY GOOD
POOR
GOOD
POOR
POOR
BRIGHTNESS
GOOD
GOOD
GOOD
FAIR
GOOD
POOR
FAIR
S i I ver Ha' ide
'*
Equ i pment:
(I)
(2)
(3)
(4)
Aeronutronic
TRW DODDAC FuJI Color System
TRW Integrated Hulticolor Display Console
GE Efdophor
(5)
(6)
(7)
A
International Electric Corporation
RCA
Fenske, Fedrik ~ Hiller Iconorama
FIGURE 5-7
COMPARATIVE CHARACTERISTICS OF SELECTED DISPLAY SYSTEMS
plus 6
Ha I
ide
** Glass Slides
Often
Need
**
to be Changed
~
----- _._----_.-_........--- ..~........ ..
---.. -
_.-----_ ..• _-_._----_ ... _-- ..._.... -.
--.-.----.--~---------~--------.--
P3958-0 79
Page 5-34
5.5.3
User and Programming Techniques (Continued)
File System.
With such programming systems it is only
nec~ssary
to append
specialized formBttlng routines whenever display systems are appended to the
computer.
There ere other problems however whIch the system
consider.
plann~rs
These are we11 illustrated in the paper on the DODDAC
the complete system documentation of the DODDAC display?
must
Sy5t~)
and
In particular
the system planner must recogniE8 and choose among the hardware/software
trade~offs
In the display/computer interface and with respect to the degree
of automatization desired.
Amongst such items are:
a.}
Are displays physically created, stored, and retrieved or are
b.)
they computer generated each time thEY are accessed?
In what way are displays requested? What flexibi II ty in format
wf 11 there be?
c.}
What feedback signals should there be from the display system
to the computer such that the entire system is integrated?
d.)
What are the timing limitations, as, for example, in driving
a fine drawing servo?
e.}
Shall backgroonds be digitized or on accessible hard copy?
Finally it wi J 1 be desirable to present some typical configurations
and indicate
th~jr
abl Ifty to sQtisfy the kinds of parameters system designers
wi I' be expected to develop such as expected frequency of requests, expected
response times, type of characters or symbols desIred, amount or density of
information to be supplied on
individu~l
presentations, etc.
*See References (9) and (10).
._
..•..
__ __ - . - - - - - - - - - - •.
..
•I
•I
P9958-079
Page 5"'35
Hardware Techniqyes
5.5.4
Many of the hardware techniques for use in console displays
described in Seclion 5.4.4 are applicable to group displays.
However,
group displays hdve the added requirement of bright, large screen
images.
Techniques such as direct cathode ray tube viewing are,
lherefore, not applicable.
To date, projection systems using photo'"
graphic images made from cathode ray tube exposures have been more
widely used for gruup displays.
A study by the Rome Air
Developn~nt
Center, "Criteria for Group Displays Chains for 1962-65 Time Period"
(ASTIA 283390), recomn~nds film base systems for group displays at
least through 1965.
For a Naval Tactical Data System for the 1970-80
era, it will be necessary to determine whether film-based systems
will still be the best or whether newer techniques such as the light
valve projector, photochromic storage, electroluminescenL screens,
or other large screen display techniques will be feasible and more
desirable.
For optical projection systems, it is necessary to choose between
techniques such as
conven~ional
si lver huJide fi Jm, pholochromic
film, Kalvar and other Diazo maleridls, zeographic processes, and
Lhermo-p J as tic record j n9.
Ea(..h of
~hese
tech!, i ques of fe r5 some
advanlages that mU5t be balanced against i
l~
5hortcomings wi th
respect to the requirements of the specific appl ication.
the display techniques investigated to
d~te
disadvantages have been covered in the
~roceedings
on dis play 5 y sterns (11, 12,
J3)
and their
Most of
adv~ntages
and
of three symposia
.
•I
•I
------------------- ---------------- . _ - - -
r -------- - . ---.-
P9958-079
\
Page 5-36
I
5.6
COMMUNICATIONS DATA HANDLING
5.6. 1
Reguirements, Systems and Uses
Today's tactical command and control systems are largely
communications handling systems.
The emphasis on communications handling
wi 11 continue; it wi 1 1 be an essential ingredient of future systems.
In a general sense the data to be received by command and control
computer systems can be divided Into two types, structured data and
unstructured data.
Structured data refers to data which Is highly formatted
and arrives at the computer at a known rate, known word structure and with
a pre-determined meaning of the digits.
Unstructured data comes to the
computer for example as English language paragraphs, or as data implicitly
recorded in photographs.
One of the prime examples of unstructured data
is intelligence information although some high level command information
might be unstructured as well.
It is likely that most of the Information
handled in tactical data systems wi 11 be structured.
However, careful
consideration must be given to unstructured information for there wi 11
undoubtedly be some of this to be handled.
Structured data Is by far the easiest to handle.
Decisions need
to be made on the equipment to perform the logic and electronic functions,
and consideration must be given to timing and synchronization.
Unstructured
data needs decisions by people before it can be placed into the computer.
This decision process can be aided by on-line consoles where the computer
aids the man In the structuring process, "te11Ing" him how this is to be
done step wise.
or
The man can then extract the Information for the computer
correct garbling, and the like.
Much of the communications problem Immediately relates to the
general Input/output problem which is discussed elsewhere.
There are,
however, a number of operations which take place:
•
I
•
I
-_ ...
I
I
I
II
-_ _.__._---- ---_._._.- -------~
..
P9958-079
Page 5-37
5.6.1
Requirements,
Sy~tems
and Uses (Continued)
1.
Information is received at a communications terminal device.
2.
Itis the-n multiplexed Into the computer system probably
simultaneously with other communicating devices.
3.
The information is then buffered unti I it is ready to be
entered for use by the computer.
4.
The message is then decoded and serviced.
5.
Very frequently the
i tse 1f
6.
mess~ge
is logged and then the message
is reco'rded.
Concerning the output of data, an output message is composed
and forma t ted and forwa rded to a proper dev i ce.
7.
After the usual operations of multiplexing and buffering, the
message is finally transmitted.
Often, of course, where the message is received on a short cycle basis
such as the receipt of radar data every 100 mi 11isecond, there is no need
for certain operations such as logging.
Communications in systems have many functions.
Some of these
are listed as fol lows:
1.
Due to physical and topographical requirments, information is
required to be transmitted from one position to another.
An
example of this is radar horizon limitations which limit the
placing of radars.
2.
Frequently it is desirable to have redundant cOmputers for
re 1 lab j 1 j ty -- tha tis, I n case one computer becomes I nope rat ive;
thus requirements forcommunfcations are generated.
3.
Frequently it is desirable to have communications from computer
to -'computer to allow a smoothing of the workload of the total
system of computers.
4.
It is usually necessary to use communications for centralization
of the command and to coordinate the entire tactical command and
control system.
I
l ______ ._.... ______________ .___ .__... ___.... _. __._.
.__ _
•I
•I
I
_ _ _ -.JI
P9958-079
Page 5-38
Terminal, Multiplexing. Storage and Buffering Equipment
5.6.2
One of the unresolved problems that has arisen with the use of
tactical comm.nd and control systems has been the determination of the
equipment configuration to perform the interface functions of terminal,
multiplexing, storage and buffering for communications and real-time
inputs and outputs.
Figure S-8
is a schematic of these Interface
functions.
At one extremity it is recognized that terminal equipment is
required to perform the terminal functions of digitizing and grouping of
information.
At the other end the interface to the computer requires that
there should be a suitable electrical Interface and that grouping of data
be performed that is acceptable to the computer.
However, there are a
number of ways for performing the functions of multiplexing, storage and
buffering.
The various technqiues are described in the following paragraphs.
F i gu re 5 - 9
is a schema tic of the Sa tu rn Stage I I checkou t
system which is representative of the phi losophy that requires that the
computer perform the bulk of the work for the interface functIons.
the system requires that the programmer program every function.
Simply
A very
rapid and large amount of information, such as is developed when either the
flight control or engine test sequences Bre exercieed, requires that the
computer be dedicated to thIs function only.
control the storage and buffering.
been built
BS
pnrt of the equipment.
The program is required to
The multiplexing in
th~se
sequences has
For less critical functions, the
programmer has to perform the multiplexing by using a 5et of special addresses
in the desired sequence.
Figure 5-10
is a schematic of the PMR real-time data handling
buffer which is representative of the philosophy that reduces the workload
to the computer to a minimum.
Thes~
sy~tems
re~~ive
the data automatically
and perform the functions of Multiplexing storage and speed changing.
At
the end of a complete data frame the cOMputer is fnformed that a complete
dat~
i
frame is now available.
L ____________________________________________________________________
•I
I•
~
~
~
TERMINAL
UNIT
STORAGE & BUFFERING
MULTIPLEXING
TERMINAl
--
TERMINAL
UNIT
........
TERMINAL
UNiT
........
..........
....
.....
SCANNER
......
......
....
-
ASSEMBLE
STORAGE
COMPUTER
-- INPUT
-
.....
.....
AND
COMPUTER
&
OUTPUT
MULTIPLEXOR
~
......
~
TERMINAL
UNIT
----
--
.....
......
ASSEMBLE
STORAGE
FI GURE 5-"8
SCHEMATIC OF INTERFACE FUNCTIONS
.......-
--
P9958-079
Page 5-40
STAGE
Ground
t---~
Automat I
Support ~-----4 Checkout
II
CDC
Equipment
~--~
Equipmen
t - - - - -__~
924A
Computer
FIGURE 5-9
TYPICAL SATURN V STAGE II CHECKOUT CONFIGURATION
.....
-- -....
COLLINS
KINEPLEX
BUFFER,
SYNCHRONIZER
and
--
.....
-
CP 6428/USQ.-20
COMPUTER
REFORMATTER
-..... -....
COLLINS
KINEPLEX
FIGURE 5-10
TYPICAL PMR REAL TIME DATA HANDLlNG BUFFER CONFIGURATION
L - - -_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ._______ . _______________ . _____________ . _ _ ._ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .___. _ .
r-----------· ------
..,
. _ - - - - - - - - - - .._---_._. __ . _ - P9958-079
Page 5-41
5.6.2
Terminal, Multiplexing, Storage and Buffering Eguipm~nt (Continued)
For shipboard command and control
missions are required.
However, it
is more
optimum use of
a.)
5y~tems
simi lar types of
A problem arises as to which approach is best.
probable that the approaches wi II be compromised for
resource~.
This factor has two aspects, the first is the
Cost actual dollars involved, and the second Is the time expended
to develop, design and implement a system.
b.)
Reliability
The concept of reliability enters in the desire
to make the interface equfpment as trouble-free as possible.
In general the simpler the functions the simpler the maintenance.
c.)
ComputE'r Capacit;,: -
Although it is possIble to assign to the
computer the interface function, an important factor is to
consider the basic workload of the computer.
Wi 11 the incremental
addition of interface function seriously reduce the computer
capacity reserve?
In the PMR example the Buffer, Synchronizer
and Reformatter were Implemented because of the effect on computer
capacity reserve.Of course if the reserve is exceeded, then special
equipment and/or computer capacity has to be increased.
d.)
frogrammer Workload -
Related to computer capacity is also the
problem of computer programming workload.
In the Saturn example,
the programming cost has become a sf9nificant factor.
Programming
resources are relatively scarce so that some considerations to
husbanding these resources are important.
e.)
Flexibi lity -
In developing, designing and implementing interface
functions there arises the need for frequent changes in requirements·;
such as Input-output characteristics of timing volume, electrical
characteristics, avaf lable information about input characteristics,
and a host of other changeable specifications.
spec~lized
The use of
gear has to be considered as compared to the general
•
•
i
I
L
_ _ _. _. . __ . . _______.____._.______.___ .___.__ . _. _._. __. _._ .___._ _._ .__. ____ .~_.
I
I
_______________.______.-J
I
1----· ---------.-----------.-.-----
-.-
------_._---
-----
I
P9958-079
I
I
Page 5-42
5.6.2
Terminal. Multiplexing. Storage and Buffering Equipment (Continued)
purpose computer.
The choice is not clear, especially if the
special gear has plugboards that are well designed to anticipate
changes.
The above cited factors indicate the extent of the problem and
indicate the areas in which further Investigation should be
5.6.3
m~de.
Pro9r8mmi~Techniques
A specific programming subsystem is contemplated for handl ing
the COO1mun i cat J ons da ta for both recept i on and trans: I j ss ion.
subsystem is in reality a kind of store-and-forward
~y5tom
Su~h
a
from a func.tional
point of view.
Three kinds of programming pieces must be identified.
are the special input/output
routin~s,
First there
including the programs permitting
flow of data into and out of the computer.
Then there are the message
management programs which control buffer size, transmission checks, error
detection and correction routines, format control etc.
Final Iy,there are
the programs that deal with content itself, determining the nature of the
message and their disposition.
Actual operation with the data is relegated
to processing programs.
From the point of view of the system designer this programming
system is very much machine dependent for the first set.
It is communication
process dependent for the second, and is message oriented in the third.
Always,however, the programs wi 11 be specified modularly -- so that new
peripherals or new messages can be added without disrupting the entire system.
Among the processes that will be studied and specified are:
Redundancy checks
Field val idation
Duplicate searching
Batch or single entry processing
Correction vs. re-transmission economics
•I
•I
I
____________J
r--··---··------·----·----··-- .- . ----.-.-.-----------.--------.----..-.. --..- - . - - - - - - - .
P9958-079
I
Page 5-43
I
5.6.3
Programming Techniques (Continued)
Exception file processing
Message logging and accounting
Dis t r I bu t i on
Of considerable influence in this endeavor wi 11 be the efforts
already InitIated by the Navy to Improve the JOPREP message system in
addition to its own fleet and logistics reporting
proc~dures.
The current
efforts on AMPS at Ft. Ritchie is another important area which will influence
such considerations.
The result of the programming system design and selection of
techniques will then afford design parameters in terms of required buffer
sizes, processing times and, as a consequence, message throughput rates.
Hence if given the input loads, it wi II be possible to come up with a "best"
system concept where trade-offs may be made in computer memory size, speed,
auxiliary storage capabi Ijty and computer organization.
important to compare and
evaJu~te
It is, for example.
the single processor approach versus the
de-centralized system, especially for the input process where a specific
computer may be delegated the communication function.
Such systems are
becoming more pronounced as evidenced, for example, by the IBM 774U
I
I
I
Ii
I
L
__.______ .________._. ___________. . _..___.__ . _... ___._______ .______ . . ___ .___.___ . ___._______ . __.__ .___. _. ___._.__.- . -__ _
and
"(~
17~U
.
•I
•I
---~
----------,
I
I
I
I
I
P9958-079
Page 5-44
REFERENCES TO SECTION 5
10
Evolution of Digi tal Magnetic Tape Systems for Use in Military
Environments, O. H. Tyrrell, D. J. Morrison, and J. J. Staller,
PROCEEDINGS, Fall Joint Computer Conference, November, 1963.
2.
Introduction to Navy Tactical Data Systems, G. W. Oi ttmann,
National Convention on Military Electronics, Washington, D. C.,
September 11, 1963.
3.
Minutes of the ASA Committee X3.1 (Optical Character Recogni tlon)
and its Subcommittees, American Standards Association, Sectional
Committee X3 on Computers and Data Processing.
4.
Electronic Retina Character Reader, DATAMATION, 9, 7, July, 1963, p. 50.
5.
Design of a Photo Interpretation Automaton, W. S. Holmes, H. R.
Leland, and G. E. Richmon, PROCEEDINGS, Fall Joint Computer Conference, December 1962.
6.
ARIOe Displays, R. T. Loewe, Electronic Information Display Systems
Spartan Books Inc., Washington, D. C., 1963, pp 231-246.
7.
Photochromic Dynamic Display, E. J. Haley, Electronic Informatioo
Display Systems, Spartan Books, Washington O. C., 1963, pp 110-120.
8.
Display Applications of Electro-luminescence, M. S. Wasserman,
Electronic Information Display Systems, Spartan Books, Washington,
O. C., I 963, pP I 2 I - I 28 •
9.
DODOAC - An Integrated System for Data Processing Interrogation
and Display, W. F. Bauer and W. L. Frank, PROCEEDINGS, Eastern
Joint Computer Conference, Washington, D. C., December ]961.
10.
OOODAC, Advanced Operational System" Final Design Report, Ramo
Wooldridge Division of Thompson Ramo Wooldridge Corporation,
Contract DA-49-146-Xl-103, Report CI53-2S-30, Vol. I and I I,
classified SECRET.
•
•
I
I
('~'-'-----~"
.---.-~
----.-..- - - ._-_.. _-- ..--,------ ..--.- ....
-~.
---.-.------ ---_ __ .._...._._ .. ._-_.. _.- ---...
'-'
.---.--.~.----.-
...
---------.-------
P9958-079
Page
R~FERENCES
11.
5-4~
TO SECTION 5 (Continued)
PROCEEDINGS) First National Symposium on Informatfon Display,
March 14, 1963, Society for Information Display, Los Ange1es,
Cal 'forni a.
12.
PROCEEDINGS, Second National Symposium on Information Display,
October 3-4, 1963.
13.
Electronic Information Display Systems edited by J. H. Howard,
Spartan Books Inc-, 1963.
•I
I•
__ ._._ .. .. _-l
P9958-079
Page 6-1
6.
HARDWARE TECHNIQUES
GENERAL
6. I
New and improved components and hardware techniques will play an
importanl role in obtaining greater capabi lilY, higher rei iabi I i ty,
and easier maintaindbili ty in compulers dnd daLd systems for use in lhe
1970-75 area.
New techniques that have already been proven feasible
wi 11 permit significanL reduction in si,e, weight, and power consumption.
These can be divided into two major categories - (I) basic componenls ind
Lechniques; (2) inlegrdled circui ts dnd other balch - fabrication techniques.
During the course of this study both existing and future componentS
and techniques will be studied and analYLed
meri
l
LO
evaluate their relaLive
and po enlial effecL on the computer field.
and lime of aVdilabl Ii ty wi II be evaluated for
cal data sYStems for which each shows
~he
Their feasibi I i ty
~hose
functions in tacli-
greatesl promise.
The advanlages
and disadvantages of different types of existing and future components
and techniques will be related to the requirements of the future Navy
Tactical Data System.
Recommendations wi II be made as to which ones
offer the greatest promise in specific types of uses in the planned
sys tem.
Existing components and techniques are capable of providing sufficient speed and performance capabil ity for the planned tactical data
system.
Ho.vever, significant reductions are necessary and possible in
the size, weight, maintainability, relial:llity, serviceability, and
logistics
requireme~ts
data systems.
for digital equipment used In shlpborne tactical
In this study, particular emphasis wi I I be placed on
components and techniques that offer promise for improving maintainab! I i ty, rei iabi II ty, and servl ceabi Ii ty.
•
•
I
I
-------.~------
-.-
P9958-079
Page 6-2
6.2
BASIC COMPONENTS
A large number of digital components dnd techniques hdve been
proposed and investigated during lhe last few yedrs.
It Is difficult
to evaluate the relative merit and value without a knowledgeable survey
of the slate of the art for each component or technique conducted wi thin
the framework of the specific applications in which Lhey mtgh: be used.
At present, and for the foreseeable future, semiconductor elements
offer the best combination of feasibil i ty, reI jabili ty, performance, and
cost for use as digi tal logical componen:s.
However, considerable
research and deveJopmenl effort has been devoted to cryogenic componenls
(1, 2.)~ magnetic logic components (3), tU1nel diode circuits (4,5),
kilomegacycJe circuitry (6, 7), and other more novel techniques in recent
years.
w.
B. Ittner, III, has pointed out that cryogenic techniques offer
promise In the memory area, but that their use as logical elements is
questionable(2). Tunnel diodes are practical circuit components.
As
stated by Eleclronics magazine, "Tunnel diodes have passed the stage
of being glamorous new semiconductor elements wi th great potential.
They I ve ar rived!
They are now jus t ano·, he r e lemen t that the eng i nee rs
can choose from to bui Jd faster, more reliable, or more sensi live
circuits. " (8)
Attention has been called to the problems of packaging,
interconnectors, and communications involved in the use of ki lomegacycle
circui ts;by D. J. Chesarek and others.(7)
New and radically different techniques, such as the use of lasers
and optical components, have been proposed and are being investigated
(9, 10, II).
All types of advanced componentsyand.!techniques should be
considered and evaluated for any compULer or data processing system
planned to be operational in 5 to 10 years - even though at present
semiconductor integrated circui ts appear to be the most feasible and
prom is I ng for that time span.
I tis I mpor Lan t that such cons i dera t Ion
and evaluations be conducted wi thin the framework of a knowledge and
*
Numbers In parentheses refer to the Bibliography at the end of
this section.
•I
•I
P9958-079
Page 6-3
6.2
BASIC COMPONENTS
(Con~lnued)
u~e
underSLanding of Lheir operaLional
wor Lhwh i 1e advan lage~ for Lhe spec j f
mus t be commensura le
COSL,
w j lh
jC
to ds~ure
dpP 1 i c.a t ion.
the i r feas i b iii ty and
and effect on rt:li/hich has many of these
characteristics but lacks the automatiGity which may be desired.
Executives for Multi-Computer Time Sharing
8.2.2
The recent trend toward on-line data processing systems operating
in quasi-real-time has given rise to an Increased emphasis on basic system
organization.
It has been recognized that careful consideration in the
design of a suitable control system wi II frequently lead to more fruitful
results in terms of speed and efficiency than wi 11 excessive expenditures
for
'1Tt i
no r i mp r ov eme n t 5 I n eq u i pmen t .
Two originslly distinct techniques have been proposed to improve
computer organization and use.
On the one hand there has been the concept
of multi-programming or time sharing of a given computer by several programs
which operate concurrently.
On the other hand there has been the multi-
processing concept in which true simultaneity of operation has been achieved
through the use of distinct but inter-connected computers.
~nd
Since command
control systems do involve multi-computers (NMCS, NTDS, OPCON future
473L and 465U executive control of the system Is important.
Historically speaking, the former approach has been adopted by
those
\~ho
have been concerned primari 1y with efficient use and
s~rvicing
of
human operators, and the latter approach by those more concerned wfth
efficient use of system equIpment.
This had led to the emergence of
divergent phi losophies of executive control.
Modern usage, hONever t indicates that no fundamental incompatibility
exists between the two techniques.
A multi·processor installation in which
individuQl computers are multi-programmed is now known to be feasible.
It
is still true that such systems are generally regarded as either a multiprogrammed "computer" having more than one Illogic unit" or as a multi-
•I
I•
I
I
.-..--_ ... _........... ___ . __ ._._________.J
r------------------.----------------
------·_---------------------------------1
I
P9958-079
I
I
Page 8-9
I
i
8.2.2
Executives for Hulti-Computer Time Sharing (Continued)
computer system in which one physical "shell" can contain more than one
"computer" .
This distinction, however. becomes entirely psychological once It
is acknowledged that the true objective In either approach Is optimization
of the total man/machine system.
to def I n i ng "opt Imi zat i on". and
The placing of emphasis is equivalent
j5
determi ned dynaml ca 1 Iy by the operat i ona I
requirements of each individual Job.
Informatics Inc. is currently engaged in the development of just
such an Executive Control System under contract to the Information Processing
laboratory, Rome Air Development Center.
The primary purpose is to develop
techniques for command and control. In this particular project the situation
Is complicated by the fact that the equipment configuration on which the
system Is designed to operate consists of an assortment of devices produced
by different manufacturers representing different periods in time.
This
configuration, moreover, Is regarded as the first phase of an evolutionary
Installation.
It Is required, therefore, that the system should be, to
the greatest extent practical, Independent of the specific equipment and
configuration.
To achieve this end, the Executive Control System (ECS) Is divided
into two major levels, as shown in Figure 8-1.
An Administrative Control
Prog ram (ACP) is res i dent I n the execut ive computer (8 CDC 16-A).
ACP is
able to distinguish between various categories of equipment but is not
concerned with their individual characteristics.
A set of local Control
Programs (lCPs) are also provided, one for each programmable unit of the
complex.
The lCPs are responsible for requesting program and data transfers
to and from the unit, but cannot initiate a data transfer to an external
device.
lCP Is also responsible for exercising any required control over
purely internal functions of each computer.
I
I
L_._ .. ___ ._..._....... ___, ____ . . _ . _.... _._. __.......____.____. .___ ... _. ___ ._._. _____ .__.____ ._._ ... __ ... ___ _
•
•
Every computer in the complex
I
I
I
I
__ __.._._...____._____ ._J
ACP
LCP
LCP
JOB
JOB
JOB
JOB
LCP
JOB
JOB
JOB
-0
OJ
-0
CD
('I)
(J1
(X)
to
CD
(X)
I
I
0
'-01
0
FIGURE 8-1
EXECUTIVE CONTROL SYSTEM HIERARCHY
to
P9958-079
Page 8.. 11
8.2.2
Executives for Multi-Computer Time Sharing (Contrnued)
will be multi-progranned, at least to the extent of being time shared by
one job program and Its own lCP.
Where more than one job program may be
operating within a computer, the complexity of that LCP wi 11 be greater
but the ACP will not be directly affected.
Executive programs for multi-computer systems have been developed
for the RW-400 computer originally intended for the Air Force 117L
program, and for the OPCON Center for CINCPAC.
The latter uti llzes
approximately five CDC-1604 and two CDC-160 computers.
One of the most prominent executive control programs for time
sharing has been that developed by SOC on the ARPA project.
It emphasi%es
the use of the computer "slmultaneously" by many analY$ts on an on-line
basis.
8.3
PROGRAMMING MANAGEMENT
Utility and executive functions were discussed in the first three
parts of Section 8.
It will also be pertinent to this study project to
structure the programming environment within which large real-time systems
evolve, and to supply an insight into the problems that must be considered.
Also It wi 11 be of help to system planners to have at their disposal some
yardsticks by which measurements can be made regarding size and time
requirements of the job Itself.
Examples of large systems which apply to such a consideration are
473L. 46Sl, SAGE, NTDS, OPCON and NMCS.
Hence, it wi 11 be desirable to
discuss the programming management with such groups. A consequence of this
investigatfon wi 11 be management recommendations. examples of which are
(2) and (3).
In addition. tables wIll be supplied reflecting pertinent
parameters measuring the programming process.
A start has been made in
doing this and it is possible to obtain data for large systems such as
shown In Figure 8-2.
can be deduced.
It is from this kind of information that guidellnestt
For example. figures relating to the cost per checked-
I
III
I
I
I
I
I
I
i
i -_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _._______________________.. ______ ... _ _ _ _ _ _ _ _ __
L
tt
III
-----_ _ _ -_0_-__- _ - - - - - - - - -
P9958-079
Page 8-12
~8.3
PROGRAMMING MANAGEMENT (Continued)
out Instruction are useful. These figures generally range in the $10-$15
range but this depends on the amount of system design and analysis included.
Maj. Gen. Terhune, Commander of ESO, USAF, recently made statements that
Air Force L programs costs for programming Is $32 per fnstruction.
He also
said that costs Increase according to the square of the number of instructions .
I
L__________________________
•I
•I
P9958-079
Page 8-13
Calendar Period - 24 months
Total Technlcl:Il Personnel, Man/Month
2,032
Total Computer Hours
6,010
No. Instructions Developed
373,500
Pages of Final Documentation
4,500
Computer Time Used/Programmed Instruction
Computer Time Used/Man Months
.965 min.
2.96 hours
Programmed 'nstructions Yielding One Page of
Final Documentation
Prog. Instruction Yie1d per Man/Month
Flnel Documentation Pages/Man/Month
85.6
184.0
2. 1
No. of Besic Program Systems
13.0
No. of Major Computer Installations Used
14.0
No. of Geographic Locations Used
8.0
Compu ters Used . . . 7090, 1604. 140 I, 160
Program Language Used . . . JOVIAL and Machine Language
Total P8ges of Interim Documentation
20,000
FIGURE 8-2
STATISTICS rOR IMPLEMENTING ONE COMMAND AND CONTROL SYSTEM
•I
•I
P9958-079
Page 8-14
REFERENCES TO SECTION 8
I.
"An Approach to On-Line Proces5ing", S. Blumenthal,
DATAMATION,
2.
June
1961.
"Management Techn iques for Rea I-Time Computer Programml "g"
T. Holdlman, JOURNAL OF THE ACM, July 1962.
3.
"Programmlng On-Line Systems", W. L. Frank, et al .•
DATAMATION, May and June 1963.
•I
I•
---------------------
P9958-079
Page 9-1
9.
ADVANCED USAGE TECHNIQUES
A number of interesting and promising techniques of computer
usage are, as yet, in the experimental stage.
Certain of these, whi Ie
showing no current appl ieabi Iity to command and control technology, are
promising for future use.
9.1
LEARNING AND SELF-DIAGNOSING TECHNIQUES
The techniques of learning and self-diagnosis, which are related,
can be said to be in the experimental stage today.
distinguish the attempts in this direction:
Two approaches
first is the building of
special-purpose machines aimed at having learning capabi tities.
Such
machines fall more or less in the classical category of "robots " (1).
Second are attempts to program general purpose machines to exhibit
structures of learning
~n
(2).
particular the work of Newell, Shaw, and Simon at the RAND
Corporation, Minsky, Selfridge, and others at MIT, and of investigators
at the Bell Laboratories and the System Development Corporation has
received notice in the technical and secular press.
Most work of
investigators seeking to build or to program a machine exhibiting
learning has been addressed to the limited areas of game playing and
theory, pattern and sound (visual and aUdio) recogni tion, including
speech, and simulatron of various simple kinds of human or animal
behav i or.
It is well to distinguish between those devices which are
essentially sensors (such as the IBM "Shoebox,'1 designed to translate
the spoken Arabic numerals zero through nine into unique bit patterns)
and those which involve self-generation of response patterns.
The
latter are intended to be the "Iearning" elements in the system.
Wooldridge (5) has suggested that the establishment
•I
I•
---------------------------- --------------
_ _ _----l
P9958"079
Page 9-2
9.1
LEARNING AND SELF-DIAGNOSING TECHNIQUES (Continued)
of permanent, closed neuron chains or loops in the human brain is the
basic mechanism of pattern recognition
J
memory, and, hence, of learning.
He further suggests that such mechanisms are constructible in electronic
machinery, but that the compJexi ty of such circuits Is beyond presentday technQI09Y and economy.
Some of the attempts to bui ld special-purpose
machinery make use of analogs of the neuron chain, whi Ie others do not.
The structure of simulation programs on general purpose machines, however,
does not follow such a pattern. due to the basic structure of the digital
compu ter.
The ultimate usefulness of simple learning has obvious future
application in command and control systems, since the command problem is
greatly simplified by abi lity to command a system with learning abi Jitles.
9.2
HEURISTIC PROBLEM SOLVING
Heuristic problem solving Is concerned with the information
processes of complex intellectual behavior-thinking.
Various attempts
have been made to program certain of these processes on digital computers.
To do such pro9ra~ning, much thought has been addr~5sed to understanding
this type of behavior-thinking process in human thinking. To date the
understanding is incomplete and disjointed.
So farJ no programs have been developed which give the machine.
capability to handle problems of significant intellactual content.
One key concept In heuristIc programming is effective search
reduction.
Hence such concepts as the content-addressable memory are
central to heur[stlc problem solving.
The solution spaces for any but
the most trivial of such problems are extremely large, and solution
algorithms are not known for most problems of intellectual Interest.
Heuristfcs are rule-of-thumb tricks and strategies which may
i
guide search of the so1ution space Into areas fertile with potential
•
I
L___________.__________._ _._ _ _ _ _.____._ ._______._ _ ._______. . ___. ___._______.__.
I
I
I
__IJ
_._----_._------------P99SS .. 079
Page 9-3
9.2
HEURISTfC PROBLEM SOLVING (ContJnued)
solutions, ignoring areas which are relatively steri Ie.
In spite of the magnitude of the general task, heuristic. programs
have been written for many tasks on
include:
general-purpos~
programs whic.h prove theorems in
computers.
Eu~lidean
symbolic logic, which do indefinite integration of
These
geometry and
comple~
Int_grands.
which play checkers, chess. and other games, balance assembly lines.
select stock portfolios. and so on.
Certain of the approaches of applied mathematics have Heuristic
content, such as the
N~wton-Raphson
and
rtl~t~d
iteration methods,
approximate solutions of various kinds, relaxati00 methodS of solution,
and other related search-reducing approaches.
Th~
application of Heuristic problem-solving methods to commlJnd
and control systems wi 11 undoubtedly be of irnportanc{; in the future.
9.3
LANGUAGE TRANSLATION
The applicabi l/ty of language translation to command and control
systems is difficult
computer~programmed
t~
foresee. However, some of
work involved in
th~
language translation is closely related to the two
prevfous subjects, so thBt the inclusiDn of the topic here's not
unrelated.
Further, some techniques involved in programs related to
1anguage translation are useful in command and cJntrol systems (7).
w'hi Ie language translation :;;.fforts
u~ing computers
remarkable successes, th('y are by no means economica1.
preparing input for
compLJt~r
translation sti 11
first-rate professional transJation by people,
exc~eds
have shown
The cost of
the total cost of
Computers will begin to
match the costs of human translation only with the development of a
successful high-speed optIcal print reader.
The machine approClch to languagf: translation varies, but most
•I
I•
._ _ _ _._ _ ._ . _ _. ._ _ _._._._.__ J
programs are constructed to oper2te SO'lw\vhat in the following order:
I
L._____. _. .___._ . . ____. _ ._. . __ . __.. . _. . . _ ._. _____._ . . __._ _. . _______. __._._.
I
P9958-079
Page 9-4
9.3
LANGUAGE TRANSLATION
(Continued)
1.
Read a block of input text.
2.
3.
Dictionary look-up of each word.
List homographs and words missing from dictionary.
4.
Assign grammar code to above words where possible.
5.
6.
Syntactic analysis of sentences.
Resolve multiple meanings, choosing correct equivalents.
7.
Output translated text.
While machine translations are, by and large, not elegant at
the present state of the art, they often yield a surprisingly complete
trans latlon.
It is in the areas of those routines which perform the dictionary
function and the syntetic analysis, and those which resolve multiple
meanings that one may expect to find techniques applicable in command
a nd con t ro I .
Work in translation of languages, notable Russian to English, is
being done at a number of locations, and is quite fully reported in the
technical literature.
9.4
APPLICATIONS IN FUTURE COMMAND AND CONTROL SYSTEMS
It will be many years before most of the learning. heuristic, and
language translation techniques are useful in everyday command and
control prc)blems; they are essentially research techniques at present,
with limited practical usage. However, it Is important for the work
proposed to understand thase techniques and extrapolate them 2-5 years
in the future or longer.
aspects of these systems
Also, It Is Important to realize that many
~
become important for practical systems even
though the technique as now conceived and developed cannot be. For
example. elaborate diagnostic programs. assemblers. compilers, etc. in
the future may well be based on the idea that the program performs a
self-improvement function or Is to a degree introspective. These
capabilitIes, while modest by research standards, may be efficient
I
and pract I Cal I.
tt
I
tt
I
I
I J
IL______.______ .____________.__._ _.__________._ .__ . _ . ___.____._. _________________ . _.__._____.________.___________._________ _
r---------·-------------··-·--··-··-·---·-·---------------.-- -.-.. -.-.----.----- ---.. --.
I
P9958-079
Page 9-5
!
REFERENCES TO SECTION 9
1.
The Perceptron - An Experiment In Learning, W. E. Bushor,
ELECTRONICS, 33, pp 56-59, July 22. 1960.
2.
A Pattern Recognition Program that Generates. Evaluates,
and Adjusts its Own Operators, L. Uhr and C. Vossl.r,
3.
4.
PROCEEDINGS, W.J.C.C., Los Angeles, 1961.
Advanced Computer Applications, W. F. Bauer, D. L. Gerlogh,
and J. W. Hranholm, PROCEEDINGS of the IRE, 49,1, January 1961.
Principes d' Incertitude de 18 Perception at Machines
Philosophiques, A. A. Moles, CYBERNETICS, 2, I, pp 51-57, 1959.
5.
6.
7.
The Machinery of the Brain, D. Wooldridge, McGraw-HilI
Book Co., New York, 1963.
Attitudes Toward IntellIgent Machines, P. Anner, The Rand
Corporation, Santa Monica, P-2114, September 30, 1960.
705 Indexes Dead Sea Scrolls, COMPUTING NEWS 66,3. April 15,
1958.
•I
I•
i
i
i
!
I
I
L.____.__. _____. ____. _____._ . _._ . ____.__.__ . _. __________. _. '._. _____._. . ..
-----.--
.....
- ..- - . -..
.
----~
P9958-ll79
Page 10 ... 1
10.
10.1
COMMENTS
ON
TW~.1!.AVY..2.y..~TtM~
FUNCTIONS AND REQUIREMENTS FOR NAVAL TACTICAL DATA SYSTEMS
The existing Naval Tactical Data System is the most sophisticated
Tactical Command and Control System in operation today.
It includes not
only a multi-computer system with associated displays, storage, and
input/output equipment, but also an integrated multi-ship communication
system providing automatic digital data transr'lission of informatin stllred
in the
canputers
on one
sh i p to those of another sh i p.
In add it i on thl:"
system is closely integrated with the Air Tactical Data System and the
Marine Tactical Data System to provide an integrated command capabll i ty
for ground, sea, and air forces.
It is essent I altha t fu tu re Nava 1
Tactical Data Systems of the 1970 and 1980 era maintain these capabi lities
for a multi-computer and multi-unit operation.
The use of a multi-computer
system on each ship helps alleviate the logistic problem and provides
a much greater assurance of the proper functioning of critical
if one machine fails or malfunctions.
operati,'n~
With the proper programming
precautions and an adequate back-up, the remaining computers can assume
the critical tasks previously being handled by any computer if that computer
fai Is.
A multi-computer system also permits adjusting the computer capability
to the requirements of any size or category of vessel.
The abi lity of
different ships. aircraft, and ground forces to interchange tactical data is
essential to the proper operation of large combined task forces.
This
problem is more critical to the Navy than to any other service because of
its wide diveristy of responsibi llties and types of forces.
The present Naval Tactical Data System has been programmed
primarily to handle air defense operations to date.
modules are being proposed for ASW operations.
However, program
The operations and functions
required in executing most of the tactical responsibi lities of naval forces
wi 11 be proved by operational experience on the present NTDS system, prior
to the completion of the planning of the 1970 - 1980 tactical data system .
•I
•I
P9958-079
Page 10"'2
10.1
FUNCTIONS AND REQUIREMENTS FOR NAVAL TACTICAL DATA SYSTEMS (Cont'd)
The future Naval Tacticdl Data System should handle al I the
operations and functions planned for the current NTDS system, but with
a greater capability and flexibility and with reduced equipment size and
weight.
On many classes of ships, the available space limits the amount
of equipment of existing types that can be used.
Integrated clrcui ts and
other microminiaturiLdlion techniques will permit far greater capabil ity
in a less space to greatly enhance the overall tactical data system particularly on the smaller ships such as destroyers and destroyer escorts.
Reliability, maintainability, and serviceability will be extremely important
in future naval tactical data systems because of the critical operational
requirements imposed on the system and the difficulty of
maintaining competent technici,ans.
With the
increasi~g
trai~ln9 ·)~d
use of atomic
power on naval vessels, and the consequent significant extensions In the
time away from home bases, reductions in logistics support requirements
will be of crucial importance.
Fin a I I y, i tis very 1 i ke I y th a t the f u t u r e Nava' Tact i cal 0 a t a
System for the 1910 - 1980 era will be the center of a completely integrated
computer and data processing system.
The various electronic subsystems, such
as fire control, search radar, sonar, and communications, will be integrated
under control direction of the ships offices through the tactical data
system.
Integrated circuits, sophisticated interrupt techniques, high-
speed large-capacity internal storage and mass memories wIll permit tieing
computers in the individLial subsystems directly into the central tactical
data system and Into other subsystems, with the necessary digital data
being transferred automatically from one to another.
With this type of
system, routine manual operations that are error-prone will be minimized,
and the necessary digital data will be provided automatically and in a
timely manner for support of the human decision-making function.
To real
i~e
these requirements and capabilities in a future naval
tactical data system will require the use of many of the new components
•I
I•
P9958-079
P.3ge 10-3
10.1
FUNCTIONS AND REQUIREMENTS FOR NAVAL TACTICAL OATA SYSTEMS (Cont'd)
and techniques described In other sections of this proposal.
is most important that only those
te~hniques
and
conpon~nts
However, it
be used that
wi 11 be both feasible and avai Idblc in the requi red time frame ",Ii th proved
performance t
reI iabi Ii ty. and maintainabi I i ty.
not only that the
n~sults
Therefore, it is essential
of this study should provide sufficient Information
and insight Into the different components Clnd
determination of their feasibi lity and
techniqu~s
tim~liness)
to permit a
but also that proper
criteria be developed to aid in the selection of optimum components,
techniques or processes for each given function or requirement.
It
\o,Ji
11
be necessary to provide tho Navy planners with adequate guidelines and
criteria to aid them In determining which components. techniques, or pr,)cesses
will provide thE! best performance and reliability for the perrrtissal::le costs,
size, and weight.
These guidelines or criteria should permit the prl)pt;r
tradeoffs to be determined between speed Jnd cost and between
characterrstics (e.g. storage capacity, speed,
perfvrman~e
logical operations) etc.)
and operational characteristics (e.g. ruggedness, maintalnabl 11 ty,
rel iabf 1 i ty, etc.).
10.2
NATIONAL EMERGENCY COMMAND POST AFlOAT (NECPA)
The NECPA is a command control system application that is of
primary int.,Ptft to the Navy. it is to provide certain national command
post capabi litles for the President and the Joint Chiefs of Staff on an
emergency basis.
The functional requirements of this mission hdve already
been refined by the Joint Cheifs of Staff and have been partially
implemented for other emergency sites.
However, it is clear that the
equipment and programming of these tasks has to be modified to meet NECPA
environmental conditions.
A major probtem is the preparation of programs
existing on the CDC i604A computer for operation on the NTDS computer such
as CP667, CP642b,
UNIVAC i21S, or TRW-130 (AN/UYK-l).
There are essentially two aspects to this problem.
The first is
the brute force translation from one computer program to the same program~
I
~
I
P9958-079
Page 10-4
10.2
NATIONAL EMERGENCY COMMAND POST AFLOAT (NECPA) (Continued)
on another computer.
The solution is not easy because JOVIAL has been
used to implement part of the sites and there is no equivalent system for
programming the tasks on the NTDS computers.
The NTDS languages of CS-l and
NELIAC are not directly related to JOVIAL.
The second aspect of the problem is a study of the NTDS standards
that have to be met to determine which standards should be modified or at
least extended to meet mission requirements.
The best illustration of
this problem is in the inter-computer communication which uti lize two
interfaces. The "slow interface" is the NTDS interface which Is slow by
current standards for this function.
Also available is the "fast Interface"
which Is non-NTDS and is available at no extra cost.
There is a difference
in speed of approximately 4 to 1 between the two interfaces.
In addition,
new equipment has to be developed to meet other NECPA requirements, and
the problem arises as to whether NTDS Interface standards should be
recommended for these new equlpments. Implicit in the revision of functional
NTDS standards would be the cost and maintanance aspects which are not
trivial.
- .-_._-_._------ -- ---
- . . _.- --_.
__ _-----------
__.
..
•I
•I
._-----_._-----P9958-079
Page 11-1
II.
PROJECT
PL~
This section of our proposal includes the work st8tement, the
program schedule, the program organization and management.
11.1
WORK STATEMENT AND DELrVERY SCHEDULE
Informatics Inc. proposes to supply tne following services to
implement the subtasks 2 and 3 of the project to provide guidance to
planners of advanced Navy Command and Control Systems.
The items marked
with an asterisk will be excluded If awarded subtask 2 only.
1.
Participate with other subtask personnel in the development
of a Study Approach Plan.
Informatics personnel will travel
to and remain in Washington,
2.
D. C. to accomplish this task.
Prepare an interim Project Plan to briefly describe the Initial
work, followed by a comprehensive Project Plan describing the
technical job to be performed and flagging tschnical and
administrative problems. Tasks will be described and detal led
mi
testones wi 11 be established and schedules developed.
contents of Interim documentation wi 11 be described.
Project Plan will
3.
The
The
be updated on a monthly basis.
IdentIfy, analyze and evaluate current and projected technology
applicable to future Navy command and control systems in order
to provide Navy planners with a comprehensive documentation of
information on available technology to form a basis for their
a.)
planning decisions. Technlcal areas covered will include:
Input/output and displays,
b.)
Memory techniques,
c.}
d.)
Computer organization,
Hardware techniques,
e. )
Prog ramm i ng,
f. )
Advanced Usage techniques,
g.)
Miscellaneous areas of importance to data processing in
command and control systems.
_
- - - - - - - - - - - - - - - - - - - - - - _.. .._.-
•I
•I
- - - - - - - ---
---~---- ..
-- -- ----
-
-- .- .._---------------- - - - - - - - - - - - - - -
P9958-079
Page 11-2
11.1
WORK STATEMENT AND DELIVERY SCHEDULE
Items 3 a.) and 3 b.} will cover both uses and systems, and
~ardware.
A list of criteria will be developed to measure the applIcability of
the new developments to requirements of the system.
Insure study integration by:
a. )
Appropriately communicating with project members of other subtasks
to Insure integration of tasks with the total project effort.
*
b. )
T8kfng into consideration the various technical data developed
In subtasks 1 and 2 (viz. Development of General Systems
Requirements and Evaluation of Applicability and AvaJ lability
of Current and Projected Technology) concerning various approaches
to the system design.
In this study integration, alternate approaches wi 11 be identified,
analyzed. evaluated, and documented for use in
design.
c~nd
and control system
The relative merits of the approaches wi 11 be assessed, to guide
the planner on several approaches that he can further develop and select
from in preparing the system's Technical Development
s.
Pl~n.
Develop a system design methodology for command and control
systems appropriate to this project.
Identify, analyze, and
evaluate various techniques In system design methodology and
make appropriate recommendation and develop suitable documentation.
Discuss system simulation and methods of determining trade-offs.
6.
Develop and produce presentations and documentation on an interim
basis as well as for the completion of the project.
These efforts
wi 1 1 I nc 1ude :
a.)
Monthly typewritten progress reports in 25 copies of each.
b.)
Midway
5ubtask reports covering work completed and including an
outline of planned future work.
These reports wi 11 be supplied
for subtasks 2 and *3 approximately 6 months after authority to
proceed is given.
*
They wi 11 be submitted in 15 copies.
See Note at beginning of Work Statement.
**In the event of an award for subtask 2 only, this work statement
item refers to coordination only, not the formal Study Inte~ratlon
task of the Request for Proposal.
I
I
L._....._._ ......._____ . _. _......_. __ .. _. __._._...-.._ .... _. -.. __...._ .-........
_~_
•I
•I
...__ ...._ _ . _ .. ~.. ___ ,. __.. _. _.. _.._. ~ . ___ .. _ . __ ~_..._..J
\_._---_._._- -_
..
_-_ _._ _-_ _
..•
•..
..
...
_....
_-
_.. __ __._----_.__.__ . __ ...__.. _-_._ ... __
.
... -
_..
_.- _._..__._.. _-
---_._---_.
__
._._-._..• __._-----
P9958-079
Page 11-3
11.1
WORK STATEMENT AND DELIVERY SCHEDULE
(Methodo logy Report)
c.}
Final subtask report/covering the work performed In the
evaluation of applicability and avai lability of current and
projected technology. will be prepared in a draft form of 15
copies three weeks before the end of the contract.
After
approval by the Scientific Officer a final report wi 11 be
prepared and submitted In 50 copies at the end of the contract.
d. )
Final Integrated Study Report in draft form in 15 copies 3
weeks prior to the end of the contract.
e.)
Progress reports to document the effort In each technical area
as that work 15 completed.
f.)
Presentation of a final report material at least 2 weeks prior
to the end of the contract. to consist of an oral briefing
accomplished by visual material such as slides/fi lmstrip, or
flip charts depending upon the size of audience.
g.)
Presentations to planners groups to show the information of
the Midway Subtask Report and three additional brIefings.
These will consist of oral presentations accompanied by visual
material appropriate to the size of the audience.
Informatics
personnel wi 11 travel to Washington, D. C. to make these
presentations and the presentation of 6 f.) above.
Informatics Inc. will assign
this project as I isted in Section 11.3.
suitabl~
qualified personnel to
Substitutions by equally qualified
personnel wi 11 only be made on agreement with the Scientific Offrcer.
11.2
WORK PLAN, MILESTONES AND SCHEDULES
Figure 11-1 shows the approach by Infurr,latlcs Inc. and Its
subcontractor Hobbs Associates, to
3ch~dulin9
the
taSKS
of the project.
Hore detailed project plans will be generated shortly after the initlatlon
of the project as shown on the chart.
Informatics
*
fnc
The Project Control System used by
is dIscussed in Section 11-2.
See Note at beginning of Work Statement.
•
I
I I
•
I
I I
.__._.______J
P9958-079
Page 11 4
-
MONTHS AFTER CONTRACT INITIATION
1
).
STUOY APPROACH PLAN
2.
PROJECT CONTROL SYSTEM
PRELIMINARY PLAN
PROJECT PLAN
MOO IF ICAT I ON AND REV lEW
3.
.
5.
SYSTEM DESIGN &. METHODOLOGY
SURVEY&. IDENTIFY
DEVELOP
~
DOCUMENTATION
a)
PROGRESS REPORTS
MIDWAY REPORTS
b)
c)
SUBTASK 2 REPORT
d)
SUBTASK 3 REPORT
e)
PROGRESS REPORTS(UNDER
f)
PRESENTATION
PRESENTATIONS
9)
Internal Control Report
~
Documentation Report
8
9
11
10
13
12
L
~
~~
.. ~
~
~~
4
EVALUATE
*
7
~
-J(
STUDY INTEGRATION
ANALYZE
6
M
ANALYZE&. EVALUATE
CURRENT ~ PROJECTED TECHNOLOGY
a)
INPUT-OUTPUT AND DISPLAY
USES &. SYSTEMS
HARDWARE
b)
MEMORY TECHNIQUES
USES ~ SYSTEMS
HARDWARE
c)
COMPUTER ORGANIZATION
d)
HARDWARE TECHNIQUES
e)
PROGRAMMING
f)
ADVANCED USAGE TECHNIQUES
GENERAL AREAS
9)
~DEHTIFY,
5
L
40
6.
4
3
2
~
(2)
L~
L
3 ABOVE)
~
FIGURE 11-1
PROJECT SCHEDULE
~
~.
(2)
f 1
_._------- _._,.,,_..
__
.. _----_.--
. - - - ------- --_ ..
__ ._-----_ _.__
..
.-.-.
__ __ ._-----_.__._---------------_ ..
..
------------------------
P9958-079
Page 11-5
11.2
WORK PLAN, MILESTONES AND SCHEDULES (Continued)
Referring to Figure 11-1 the first Item to be accomplished Is
the participation with other subtask personnel In the development of a
study approach plan. This work will take place in Washington, O. C. as
specified in the RFQ and will occupy the efforts of the key members of
the project team.
Also, as mentioned above, one of the first tasks is
to develop a Project Plan.
Following this, the main technical effort of the project begins.
Each bar, representing an area of technical activity. wi II consist
normally
of the following efforts:
I.
Technologies and techniques will be identified by literature
2.
survey and by discussions with individuals.
The various technologies and techniques will be analyzed and
3.
fully understood to develop the appropriate facts for evaluation.
The technical factor will then be evalu.ted and their applicability
determined and recommendations made.
4.
The technical area will be documented.
We would like to stress the last point -- the fact that following each
effort In a technical area the work is documented so that there is a
continuity of documentation throughout the project and no vital information
is delayed.
This further enables project personnel to maintain a
continuous focus on the principal end product -- the final documentation
for the system designers' use.
Each of the task efforts as shown In the chart will be manned
by 2 - 4 project personnel.
Rather than have one person worry about
memory techniques, for examp1e, for an 11 month period, it will be
advantageous to work in teams of 2 - 4 people to appropriately exchange
ideas and to coordinate efforts.
In general, the Project Plan calls for the examination of certain
hardware techniques, the Integration of these hardware techniques into
I
I
systems, and then a full examination of their potential usefulness.
I
I
I
l______________________________ -------------------------._-----------------._-----------. ---
•I
I• ,
_______________ J
The
P9958-079
Page 11-6
11.2
WORK PLAN, MILESTONES AND SCHEDULES (Continued)
hardware activfties, It will be noted, are spaced throughout the project
that the project team members emphasizing hardware can go from one
technical area to the next.
50
11.3
PROJECT CONTROL SYSTEM
Informatics Inc. utilizes a comprehensive Project Control System
as a strict policy. The Project Manager must, after no longer than 6 weeks
following the initiation of a project, develop a comprehensive Project
Plan. Prior to this, to insure efficient work in the early weeks of the
project, a temporary Project Plan is issued, briefly describing the work.
The
Project Plan presents briefly the technical Job to be
performed and specifically refers to any technical or administrative
problems which are likely to arise. Tasks are described and teams and
project personnel are assigned to the task.
Detailed milestones and
schedules are developed. Contents of Interim documentation are described.
Dollar expenditures are planned for each task and subtask. Charts are
presented showing the rate of direct labor expenditure and the rate of
other direct cost expenditure. The Project Plan is a deliverable contract
item but the cost information will be omitted.
Throughout the life of the project the Project Plan Is updated
on a monthly basis.
The actual expenditures are presented on the charts
and compared with the planned expenditures.
Each month the Project Manager
and key project personnel make a verbal presentation to Informatics
management on the progress of the work.
The Project Control System, carefully conceived and di ligently
carried out, insures that project efforts wi 11 be efficient and economical.
There is no substitute to this approach.
11.4
THE PROJECT TEAM
The project team members and their technical areas of contributi1lt
to the project are as shown in Figure 11-2. Dr. Wa 1te r F. Baue r i S
I
•
I
NAVAL ANALYSIS GROUP
SCIENTIFIC OFFICER
w.
F. BAUER
Proj ect Manager
W. L. FRANK
Assoc.Pr
Hardware
Techniques
Memory Hardware
Input/Output
Display Hardware
Environmental
Factors
Computer
Organization
System Design
and
Methodology
Input-Output
Study Integration
Display Systems
System Design
and Methodology
Study Integration
Advanced Usage
Techniques
Progranmi n9
Memory Uses and
Systems
r----~-----,
E. COIL
Programming
Memory
Techniques
G. STOCK
Programming
Display Systems
A. ESPO
E. BURGESS
Proj ect
Coordination
Documentation
Programning
Exercise
Training
K. CURTICE
Training and
Exerci se
Programming
""0
Q)
'-C
(1)
FIGURE 11-2
--------
to
to
U'1
ex>
I
PROJECT ORGANIZATION
---------
'"'0
------- ----
I
-..I
0
"
to
----- --------------------------------,
P9958-079
Page 11- 8
11.4
THE PROJECT TEAM (Continued)
deslgnat~d
the Project H.tnager.
We recognIze the unusualness of th,ls
approach since he Is the President of Informatics Inc.
However, the
project is of great Interest to him professlona1ly and we believe that this
assignment will Insure the best administratively and technically coordInated
approach. Dr. Bauer has a long list of qualifications for this job as
described In Section 12.
Hr. Werner L. Frank Is designated the Associate Project Manager,
He lfkewlse has had extensive experience which directly qualifies hIm for
this task.
For example, he was the Project Manager on a $500,000 System
DeSign project for DODDAC which is now the National Mi1itary Command
Support Center.
Sect ion 12.
His list of qualifications is likewise recorded in
The chart shows proposed members Including Dr. Bauer and Mr. Frank.
It illustrates certain technical areas In which the personnel have special
qualifications. These areas are the likely ones for aSSignment.
Of the group of 11 personnel shown, 6 have had over 12 years
experience, each, in modern electronic computer systems t most of it
advanced military weapons records experience. (Experience in punched card
areas or general technical work is not counted.)
The remaining 5 have had
4 - 10 years experience in modern large scale computer systems.
The proposed assignments
a~d
an estimate of the amount of time
it is planned the project personnel will work on the study are as follows:
Dr. W. F. Bauer. Project Manager ";':
Werner L. Frank, Associate Project Manager
L. C. Hobbs, Hardware Techniques, Memory
30%
100%
Hardware, Input/Output and Display Hardware,
Environmental Factors
100%
H. Hess, Input/Output, Study IntegratIon
100%
J. W. G~anholm, Study Integration, Advanced
Usage Techniques·
----------------
70%
•I
•I
-----
-- ------
P9958-079
Page I 1-9
11 .4
THE PROJ EeT TEAM (Cont i nued)
R. H. Hill, Programming. Memory Uses
and Systems
100%
E. A.
100%
Coil. ProgrammIng, Memory Techniques
G. Stock, Programming, Display Systems
100%
E. Burgess, Project Coordination. Documentation,
Presentat ions
,'(
70%
A. Espo, Programming, Training, Exercise
100%
K. Curtice, Training and Exercise, Programming
100%
Note: If we are awarded Subtask 2 only, -the percentage of
time for Dr. Bauer wi 11 be negotiated.
•I
•I
L -_ _ _ _ _ _ _ _ _ _ _ _ ._____________ _
P9958-079
Page 12-1
12.0 CONTRACTOR QUALIFICATIONS
This section describes the qualifications of the companies
involved, the qualifications of the project team as a group, and the
qualifications of the individuals.
It includes an extensive bIblio-
graphy of the project team members as well as their fonnal biographical
s ke tches .
INFORMATICS INC. AND HOBBS ASSOCIATES
12.1
Informatics Inc. was organized to provide systems design
and analysis, consulting, and programming for systems involving stored
program electronic digi tal computers.
and the current projects is
mil i tary.
Adv.Jn~ed 00-1
~uch
The background of the individuals
that the work is almost
exclu~iv{!ly
i ne, or rea 1 t ir:le, conpu ter !>ys terns is Lhe
specialty.
Current total head count of Informatics Inc. is 45.
Approximately 95% of the work is invo1-./ed with the design
of on-l ine systems for mi I i tary applications.
dnd
prograrmling
This represents projects.
at the Goldstone Tracking Station, Jet Propulsion Laboratory, Pacific
Missi Ie Range, Manned Space Craft Center - Houston, N<'dional t-li I i 'dry
Command System Support Center - Washington, and the Rome Air Developmenl.
Center as major efforts.
lhe remaining projects are in programming
systems work and diagnostic programs for computers.
of key
fnform.J,jc~'
explains the
personnel as well as
jn~erest
~.hest;
The background
sj.)t!,-ifi(: I-'r
1
jf.:ClS
and qua) IfiGations of the company for lhis
projec t.
Hobbs Associates is a hardware and systems engineering firm
devoted to the study, eVu}uution, and design of digi tal equipment and
system~.
Hobbs Associates hd5 two
affiliates.
e~ployees
and :hrec professional
Specific areas of interest, experience, and capabi I i ty
include:
~,-.-----------.--.---.------
•I
•I
-
~--------
-------- ------
--
---..-- --
~----
-----
-~---
--- - - - - -- --
--
- - - - - - - - - - - --------------------------------------------------------,
P9958-079
Page 12-2
INFORMATICS INC. AND HOBBS ASSOCIATES (Continued)
12.1
Digital equipment and systems,
Special purpose and general purpose computer organization and design,
Data acquisition systems,
Input and output subsystems
Message composers and data edi ling equipment
Display subsystems
Mass Storage
Digital hardware techniques, and
Command and Control
System~.
Since Hobbs Associates does not manufaclure equipmenL, they are in a
unique position to make technical studies and
~rovide
recommendations
wi thout the danger of their being influenced by ve5Led interest in
specific
12.2
tech~iques
or types of hardware.
TEAM QUALIFICATIONS
As mentioned in Section 11.3, of the 11 proposed members of
the project team, 6 have over 12 years experience, each, in modern
electronic computer sys terns.
expe r i ence.
1.
The remaining have had 4 - 10 years
The group has the following qual iflcations:
They are exper ts in on-line computer systems -- computer
systems which interface wi lh extensive inSlrumentation
on a
2.
real time basis.
They have extensive experience wi lh mil i tnry systems
especially command and
3.
con~rol.
They know technology and techniques and dre acquainted
In detail with the data
proces~ing
industry, its products
and its people.
4.
They have pub) ished and presented numerous papers in the
technical
area~
of
the project team is
(Sec t ion 12.4).
inler~st.
pre~ented
The lJib1iugraphic 1ist for
at the end of this section
•I
I•
P9958-079
Page 12-3
12.2
TEAM QUALIFICATIONS (Continued)
5.
They have established an approach to the various task
areas as shown by this technical proposal.
Figure 12-1 shows the specific contact which team members
have had with large scale mil i tary systems related to or similar to
the Navy Tactical Command and Control Systems.
The following is a very
brief description of each of these systems.
NMCSSC - DOpPAC.
NHCSSC (National Mill tary Command System
Support Center, successor to DOODAC) Is responsible for providing basic
information on damage assessment and the status of forces and resources
during peace time as well as during war time for friendly, enemy and
neutral forces.
It is the highest level cOlTVl1and and control system
being planned in the country.
~o
The Naval Tactical Data System is the present system
for tactical corrrnand and contro].
It complemen:s or is related co
many other systems such as the Marine Tdctical Datd System and other
airborne and shipborne systems.
PMR - RTpHS.
The PHR Real Time Data Hdndl ing System is an
important system for consideration here since it employs the NTDS
computers and handles sensor informat.ion on real time
therefore includes
mdny
ba~is.
It
of the LCc:hllclogi£:s of inLt:resl in corrvnand and
con~roI.
ARTOC.
The ArrrlY Tactical
Operdtion~
Cenl€:r is the central
command post of the U. S. Field Army which wil I be implemented by
communications) compUlcrs and displays.
fIELDATA.
This is the general system of communications and
computing fuci 1 i ties for the field
i\rmy
and Includes
~uch
functions as
intelligence data hundling, weapon control, COITll:1and, dnd battlefield
surve ill ance.
•I
•I
- - - - - - - - - _..•
__._._.._-_._- .. _.__ __.__ _-----_._._..- - .
..
r - - - - - - - - - - - - - - - - - - - - - - - - - -..-------
P9958-079
Page 12-4
~
.....,......
i41
'"
~
~
§
~
~~
~ J.;;.
~ ~
BAUER
x
FRANK
x
HOBBS
p~
~ .....,J
Q:I
C(/
~ 4..
......
Source Exif Data:
File Type : PDF
File Type Extension : pdf
MIME Type : application/pdf
PDF Version : 1.3
Linearized : No
XMP Toolkit : Adobe XMP Core 4.2.1-c043 52.372728, 2009/01/18-15:56:37
Create Date : 2012:02:05 19:19:37-08:00
Modify Date : 2012:02:05 19:28:58-08:00
Metadata Date : 2012:02:05 19:28:58-08:00
Producer : Adobe Acrobat 9.5 Paper Capture Plug-in
Format : application/pdf
Document ID : uuid:35a7eb1a-2556-4172-a528-400cf86608b8
Instance ID : uuid:34d1413d-6a83-42d3-9aa4-14b9f72de550
Page Layout : SinglePage
Page Mode : UseOutlines
Page Count : 172
EXIF Metadata provided by EXIF.tools