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 : 172EXIF Metadata provided by EXIF.tools