Historical Reflections the Immortal Soul of an Old Machine Taking Apart a Book to Figure out How It Works

Total Page:16

File Type:pdf, Size:1020Kb

Historical Reflections the Immortal Soul of an Old Machine Taking Apart a Book to Figure out How It Works viewpoints VDOI:10.1145/3436249 Thomas Haigh Historical Reflections The Immortal Soul of an Old Machine Taking apart a book to figure out how it works. HE BEST BOOK ever written about IT work or the com- puter industry will be 40 years old in August. Tracy Kidder’s The Soul of a New TMachine describes the work of Data General engineers to prototype a minicomputer, codenamed “Eagle,” intended to halt the advance of the Digital Equipment Corporation’s hugely successful VAX range. It won both the Pulitzer Prize and National Book Award for non-fiction, perhaps the two highest honors available for book-length journalism. Year after year, the book continues to sell and win new fans. Developers born since it was published often credit it with shaping their career choices or help- ing them appreciate the universal as- pects of their own experiences. Soul’s appeal has endured, even though what started out as a dispatch from a fast-growing firm building a piece of the future now reads as a time capsule from a lost world. Back in 1991 1: It Does Not Assume In contrast, Soul delivers a self-con- I read the book for an undergraduate You Know Anything tained package, containing everything class, typing my paper on a PC that was Paradoxically, the obscurity of Data Gen- you need to enjoy the story. Back in 1981 already more capable than Eagle yet eral helps to explain the book’s endur- most potential readers had never used a cost 100 times less. So why are so many ing power. My shelves are full of books computer of any kind, still less a “super people still excited to relive the cre- about Microsoft, Apple, Netscape, and minicomputer.” Kidder would have had ation of a pitifully obsolete computer, Oracle written while the companies to self-publish any book written for designed by a team of obscure engi- were famous. Their authors assumed Data General fans, and anyway he knew neers for a long-forgotten company anyone who picked the book up was nothing about computers when he ar- that never mattered very much anyway? already fascinated with the company, rived in Westborough, MA, to follow up Having spent almost 30 years now try- cared deeply about its products, and a suggestion from his editor. Kidder’s ing to take the book apart and figure would enjoy endless pages of gossip, only previous book was about a murder out how it works, I think I have some corporate strategy, legal maneuverings, and his main life experience, other answers. Ten of them, in fact. and trivia. They have not aged well. than a Harvard degree in English and ASSOCIATES ANDRIJ BORYS BY IMAGE 32 COMMUNICATIONS OF THE ACM | JANUARY 2021 | VOL. 64 | NO. 1 viewpoints an MFA from Iowa, was having spent 80960, and Itanium) and even DEC’s two unhappy but uneventful years as much-admired Alpha processor. First Lieutenant, Military Intelligence Many readers The clean sheet approach failed for in Vietnam. were tantalized by Data General too. Anticipating this, Tom West rounded up the best of the 2: It Reads Like a Classic the idea of computer engineers left behind in Massachu- American Novel architecture as setts to launch a semi-clandestine ef- The book was a milestone in the devel- a creative medium fort to produce a 32-bit extension of V opment of what is now called “literary the existing 16-bit Eclipse minicom- nonfiction.” To keep us turning pages, in which experts puter, preserving compatibility by in- Kidder draws deeply on the mythic ar- terleaving old and new instructions chetypes of American literature, as in his could read traces seamlessly rather than using a “mode introduction of Tom West, his protago- of individual bit” to enter a separate legacy mode. nist and the project’s leader, during a (More than 20 years later, AMD played prologue that recalls Ernest Hemming- flare or, as here, a comparable trick on Intel by extend- way: West awes the other crew members a conservative ing the standard x86 architecture to of a small sailing boat with his stamina 64-bits). For the project to be approved and ruggedly taciturn optimism when organizational as “insurance” against problems with hit by a storm. “Whatever he did for a culture. the North Carolina team, West had to living,” they conclude, “it was probably promise the impossible task of pro- interesting and obviously important.” ducing the entire computer in one year. Having hooked us on the enigmatic Thirty engineers crowded together, Tom West, Kidder is cocky enough to turning the basement of Data General’s spend an entire chapter without men- headquarters into a site of relentlessly tioning him, instead introducing Data Smithsonian Observatory. That doesn’t hard work. General as “the Darth Vader of the com- quite justify the build-up, but whatever: The team’s outsider triumph gives puter industry” (a reference that un- we are already hooked. Confessing to way to an unexpectedly mournful con- doubtedly aged better than he expect- his editor that he was having difficulty clusion. West himself is banished to a ed). Data General’s corporate culture capturing West, “whose special vanity marketing job in Japan. Kidder, wink- was defined in equal parts by thrift and had been to make himself mysterious ing at his name, compares West to a aggression. Kidder confides that Data to me as well as to his team of computer gunslinger, who dispatches the bad General’s spartan corporate offices were engineers,” Kidder had been advised to guys only to be run out of town by the engineered for rapid conversion to fac- “do a Gatsby on him.”4 very citizens he saved. “It was a sum- tory space even before he gets around to mer romance,” realizes West. “None of mentioning the lawsuit a rival firm filed 3: It Roots for the Underdogs it came out the way he had imagined it to accuse Data General of burning down Gripping stories often ask us to root would, but it was over and he was glad.” its factory. for underdogs to triumph against the When West reappears, Kidder stim- odds. Data General’s brutal corporate 4: It Captures the “Crunch” ulates our curiosity by presenting him culture gave West space to launch of Startup Development as a figure of mystery to his own team his project but deprived his team of Although Data General was a mature members: a CIA agent, a folksinger, a resources, leaving them in cramped company, the project was run more speed freak, even “a prince of dark- and uncomfortable conditions. Even like a startup. West and his lieuten- ness.” I was struck by the similarities pencils were in short supply. If you are ants staffed the team with young men with the technique F. Scott Fitzgerald thinking that a major minicomputer fresh from engineering school, lured used in The Great Gatsby to introduce firm ought to have been able to provide with the prospect of being able to de- his title character. Jay Gatsby, urbane pencils to the elite team building its sign a new computer architecture. host of fabulously swanky parties, next-generation system, you’d be right. They boasted of being “a place where turned out (spoiler alert) to be plain However, that team and its pencils people are really doing the next thing” old James Gatz of North Dakota, a were down in North Carolina, design- but cautioned that “there’s a lot of fast lovelorn bootlegger desperate to re- ing an ambitious all-new 32-bit archi- people in this group … a real hard job capture the attention of his lost sweet- tecture on a clean sheet of paper. Data with a lot of long hours.” In short: “tell heart. Less dramatically, corporate General had caught a bad case of what him that we only let in the best. Then computer engineer Tom West, turns Fred Brooks called the “second system we let him in.” It’s a classic example out to be Joseph Thomas West III, an syndrome.”4 The tactic of throwing of what software engineering writer engineer from a privileged back- out backward compatibility for a new Ed Yourdon called the “marine corps” ground who came late to corporate architecture worked for IBM with its justification for a “death march proj- life after taking a year off during col- legendary System/360 gamble, but has ect.”7 As Kidder put it, “It was kind of lege to play folk music, followed by failed far more often, for example with like recruiting for a suicide mission. seven slightly offbeat years building IBM’s own Future Systems project, at You’re gonna die, but you’re gonna die and delivering digital clocks for the least three times for Intel (iAPX 432, in glory.” JANUARY 2021 | VOL. 64 | NO. 1 | COMMUNICATIONS OF THE ACM 33 viewpoints Eagle is brought to life more slowly their own souls accumulated to bring ing a part of the machine or for run- than the team had promised but sooner the new machine to life. ning the debugging process, intro- than Data General had any right to ex- ducing a rich cast of clearly delineated pect. Inexperienced recruits were ma- 5: It Gets to Technology supporting characters with their own nipulated into “signing up” to aggres- Through People quirks and motivations. sive schedules, because an unreasonable How to make a reader commit to al- One such chapter, “The Case of the commitment given freely motivates most 300 pages of finely observed Missing NAND Gate,” begins by intro- more deeply than one imposed by man- business history of a company they ducing several engineers.
Recommended publications
  • A Screen Oriented Simulator for a DEC PDP-8 Computer
    University of Wollongong Research Online Department of Computing Science Working Faculty of Engineering and Information Paper Series Sciences 1983 A screen oriented simulator for a DEC PDP-8 computer Neil Gray University of Wollongong, [email protected] Follow this and additional works at: https://ro.uow.edu.au/compsciwp Recommended Citation Gray, Neil, A screen oriented simulator for a DEC PDP-8 computer, Department of Computing Science, University of Wollongong, Working Paper 83-2, 1983, 65p. https://ro.uow.edu.au/compsciwp/69 Research Online is the open access institutional repository for the University of Wollongong. For further information contact the UOW Library: [email protected] THE UNIVERSITY OF WOlLONGONG DEPARTMENT OF COMPUTING SCIENCE A SCREEN ORIENTED SIMULATOR FOR A DEC PDP-8 COMPUTER .". N.A.B. Gray Department of Computing Science University of Wollongong Preprlnt No 83-2 January 25. 1983 P.O. Box 1144. WOLLONGONG. N.S.W. AUSTRALIA telephone (042)-282-981 telex AA29022 A Screen Oriented Simulator for a DEC PDP-8 computer. N.A.B. Gray. Department of Computing Science. University of Wollongong. PO Box 1144. WOllongong NSW 2500. Austr"1lia. ABSTRACT This note describes a simulator for the DEC PDP-8 computer. The simulator is intended as an aid tor students starting to learn assemDly language programming. It utilises the simple graphIcs capaDilities of the terminals in the department's laboratories to present. on the termI­ nal screen. a view of the operations of the simulated computer. The complete system comprises two versions at me program tor simulating a PDP-8 computer and a simplified "assembler" tor prepar­ Ing students' programs for execution.
    [Show full text]
  • Go Forth with TTL !
    Go Forth with TTL ! The Gigatron TTL Color Computer Forth for a Very Unusual Processor Ken Boak SV Fig. Forth Day 2019 . In September of 1975, MOS Technology launched the 6502 at the Wescon75 Computer Conference in San Francisco. Chuck Peddle and his team had created a very lean, stripped down, small die cpu. Costing just $25, the 6502 was a fraction of the cost of its nearest competitor. At that time the Intel 8080 was $360 and the Motorola 6800 was $175 . The 6502 was clearly a disruptive usurper. 25 year old, HP Engineer, Steve Wozniak, realised that this new microprocessor would be a game-changer and went on to incorporate it into the small computer he was developing. That machine went on to become the Apple I. In 1975 7400 TTL was the “Bread and Butter” of logic design: 7400 series TTL integrated circuits were developed in the early 1960’s. Initially quite expensive so mainly used in Military and Aerospace applications. By the early 1970’s TTL had become a versatile family of standardised, low cost,. easy to use logic. Typically about $1 per device. 7400 series logic was widely used in the design of minicomputers, including the PDP-11, the Data General Nova 1200 and later models of PDP-8. TTL was a viable, faster and cheaper processing solution than the emerging 8-bit microprocessors such as MOS 6502, Intel 8080 and the Motorola 6800. Essential Reading 16-bit TTL CPU board from Data General Nova 1200 The Gigatron TTL Computer – What is it? Started as a Hackaday.io project in Spring 2017 by Marcel van Kervinck of The Hague, Netherlands.
    [Show full text]
  • Computer Architecture Anc Instruction Set Design*
    Computer architecture anc instruction set design* by P. C. ANAGNOSTOPOULOS, M. J. MICHEL, G. SOCKUT, G. M. STABLER, and A. van DAM Brown University Providence, Rhode Island INTRODUCTION The Brown University Graphics System (BUGS)1* was designed as the vehicle for performing this research. Prin­ A group of computer scientists and mathematicians at cipally, the configuration consists of an IBM S/360-67 Brown University has been engaged in the study of running the CP-67/CMS time-sharing system,10 used by computer graphics for the past eight years. During the the entire Brown University community, and a satellite course of these studies a variety of topics has been inves­ display station, as illustrated in Figure 1. This reasonably tigated, in particular, during the last few years, the use of powerful satellite configuration provides such facilities as microprogramming for implementing graphics sys­ program editing and compilation, debugging tools, and tems.2021' In early 1971, Professor Andries van Dam and most importantly, application processing power and data his associates submitted a threefold research proposal to storage. However, because of the two rather distinct the National Science Foundation. The problems to be demands placed upon the local processor, that of display investigated were: generation and general computing, and because these two capabilities could run in parallel, it was further deter­ (1). Inter-Connected Processing (ICP-ing) between a mined that the inclusion of two separate processors in the central computer and an associated satellite proces­ graphics station would be in order. In particular, the first sor, with the goal of a dynamically alterable solu­ of these processors would be of a general-purpose nature, tion to the "division of labor" problem; program while the second would be designed specifically for main­ modules would be dynamically linked in either tenance and regeneration of the display.
    [Show full text]
  • Cern Libraries, Geneva Cm-P00087609 Cern/Fc/1374
    CERN/FC/1374 CERN LIBRARIES, GENEVA Original: English 15 September, 1971 CM-P00087609 ORGANISATION EUROPÉENNE POUR LA RECHERCHE NUCLÉAIRE CERN EUROPEAN ORGANIZATION FOR NUCLEAR RESEARCH FINANCE COMMITTEE Hundred-and-fourteenth Meeting Geneva - 29 September, 1971 ADJUDICATION FOR REMOTE INPUT OUTPUT STATIONS FOR THE CERN CENTRAL COMPUTER SYSTEM The document CERN/FC/1340, Programme for Acquisition of Peripheral Equipment for the CERN Central Computer System, outlines the programme of development needed to build up a decentralized service based on the CDC 7600 computer. The first step in this programme consists of the acquisition of five or six Remote Input Output Stations. Each station will be based on a small computer which initially drives a card reader, line printer and a communications controller. Additional stations may be purchased later and some will be expanded by the addition of extra peripherals and high speed communications equipment as the demands on the decentralized service grow. The offers received from firms show the Modular One computer manufactured by Computer Technology Limited to be the one which most exactly meets the technical specification. The offer from Computer Technology Limited is a few percent more expensive than the two other possible machines for the initial configuration, but this is more than compensated for by the proven software and greater capability for ex• pansion of the Modular One Computer. The Finance Committee is requested to approve the award of the contract for an initial order of five or six basic Remote Input Output Stations from Computer Technology Limited at a price of approx• imately 182,000 Swiss Francs per station (excluding the card reader), with the possibility of later 71/286/5/e CERN/FC/1374 I.
    [Show full text]
  • Architecture and Computer Storage
    White Paper Architecture and Computer Storage July 2002 Architecture and Computer Storage 0 EMC believes the information in this publication is accurate as of its publication date. The information is subject to change without notice. The information in this publication is provided “as is.” EMC Corporation makes no representations or warranties of any kind with respect to the information in this publication, and specifically disclaims implied warranties of merchantability or fitness for a particular purpose. Use, copying, and distribution of any EMC software described in this publication requires an applicable software license. EMC2, EMC, and Symmetrix are registered trademarks and where information lives is a trademark of EMC Corporation. Other trademarks are the property of their respective owners. Copyright © 2002 EMC Corporation. All rights reserved Published in the USA H749 Architecture and Computer Storage 1 Table of Contents Architecture and Computer Storage............................................................................3 Why is architecture important for computer systems? .............................................................................................. 3 Architectures—Good and Bad .................................................................................................................................. 3 What about storage?.................................................................................................................................................. 4 Breakthrough #1 that changed
    [Show full text]
  • ALTO: a Personal Computer System Hardware Manual
    This document is for internal Xerox use only. ALTO: A Personal Computer System Hardware Manual August 1976 Abstract This manual is a revlSlon of the original description of the Alto: "Alto, A Personal Computer System." It includes a complete description of the Alto I and Alto II hardware and of the standard microcode (version 23). © Copyright 1975. 1976 by Xerox Corporation XEROX PALO ALTO RESEARCH CENTER 3333 Coyote Hill Road / Palo Alto / California 94304 This document is for internal Xerox use only. Contents 1.0 Introduction 2.0 Microprocessor 2.1 Arithmetic section 2.2 Constant Memory 2.3 Main Memory 2.4 Microprocessor control 3.0 Emulator 3.1 Standard Instruction Set 3.2 Interrupts 3.3 Augmented Instruction Set 3.4 Bootstrapping 3.5 Hardware 4.0 Display Controller 4.1 Programming Characteristics 4.2 Hardware 4.3 Display Controller Microcode 4.4 Cursor 5.0 Miscellaneous Peri pherals 5.1 Keyboard 5.2 Mouse 5.3 Keyset 5.4 Diablo Printer 5.5 Analog Board 5.6 Parity Error Detection 6.0 Disk and Controller 7.0 Ethernet 7.1 Programming Characteristics 7.2 Ethernet Interface 7.3 Ethernet Microcode 7.4 Software Initiated Boot Feature 8.0 Control RAM 8.1 RAM-Related Tasks 8.2 Processor Bus and ALU Interface 8.3 Microinstruction Bus Interface 8.4 Reset Mode Register 8.5 Standard Emulator Access 8.6 M and S Registers 9.0 Nuts and Bolts for the Microcoder 9.1 Standard Microcode Conventions 9.2 Microcode Techniques Which Need Not Be Rediscovered Appendix A Microinstruction Summary Appendix B Reserved Memory Locations Appendix C Optional Alto Peripherals 2 1.0 INTRODUCTION This document is a description of the Alto, a small personal computing system originally designed at PARCo By 'personal computer' we mean a non-shared system containing sufficient processing power, storage, and input-output capability to satisfy the computational needs of a single user.
    [Show full text]
  • The Design of a Real-Time Operating System for Experimental Psychology KENNETH J
    Behavior Research Methods & Instrumentation 1979. Vol. 11 (5).507-511 COMPUTER TECHNOLOGY The design of a real-time operating system for experimental psychology KENNETH J. BURKHARDT DepartmentofElectrical Engineering, RutgersUniversity. Piscataway, New Jersey 08854 This paper describes the structure of the operating system that was utilized in the imple­ mentation of the experimental programming language. PL/E. Although the system was originally designed to support programming of psychological experiments on a specific machine (Data General Nova), the operating system can be ported with a minimum of effort to other machines. This paper describes the internal algorithms and control structures used in the system and suggests how it can be moved to other computers. Given the ever decreasing cost of computer hardware, SOFTWARE REQUIREMENTS FOR it is becoming quite clear that the cost of implementing REAL-TIME SYSTEMS any application on a computer is greatly affected by the degree of effort required to develop system software. Software requirements for real-time applications In many small computer environments, system hardware such as experimental psychology must be examined at costs are often negligible when compared with software two levels: the level of the applications software and the costs. Thus, in order to minimize total system cost, level of the operating system software. At the applica­ it is necessary to utilize appropriate design aids for tions level, the problems of real-time programming are efficient software development. Although many such quite complex. The wide disparity between potential tools exist in conventional data processing and scientific applications makes even specification of a typical environments (for example, higher level languages such environment extremely difficult, if not impossible.
    [Show full text]
  • The Computer History Simulation Project
    The Computer History Simulation Project The Computer History Simulation Project The Computer History Simulation Project is a loose Internet-based collective of people interested in restoring historically significant computer hardware and software systems by simulation. The goal of the project is to create highly portable system simulators and to publish them as freeware on the Internet, with freely available copies of significant or representative software. Simulators SIMH is a highly portable, multi-system simulator. ● Download the latest sources for SIMH (V3.5-1 updated 15-Oct-2005 - see change log). ● Download a zip file containing Windows executables for all the SIMH simulators. The VAX and PDP-11 are compiled without Ethernet support. Versions with Ethernet support are available here. If you download the executables, you should download the source archive as well, as it contains the documentation and other supporting files. ● If your host system is Alpha/VMS, and you want Ethernet support, you need to download the VMS Pcap library and execlet here. SIMH implements simulators for: ● Data General Nova, Eclipse ● Digital Equipment Corporation PDP-1, PDP-4, PDP-7, PDP-8, PDP-9, PDP-10, PDP-11, PDP- 15, VAX ● GRI Corporation GRI-909 ● IBM 1401, 1620, 1130, System 3 ● Interdata (Perkin-Elmer) 16b and 32b systems ● Hewlett-Packard 2116, 2100, 21MX ● Honeywell H316/H516 ● MITS Altair 8800, with both 8080 and Z80 ● Royal-Mcbee LGP-30, LGP-21 ● Scientific Data Systems SDS 940 Also available is a collection of tools for manipulating simulator file formats and for cross- assembling code for the PDP-1, PDP-7, PDP-8, and PDP-11.
    [Show full text]
  • On Microprocessor-Based Computers
    Behavior Research Methods & Instrumentation 1975, Vol. 7 (2), 183-186 On microprocessor-based computers E. JAMES KEHOE, LLOYD J. FREI, ROBERT W. TAIT, and I. GORMEZANO University ofIowa, Iowa City, Iowa 52242 Economical microprocessor-based computers have been recently developed by Motorola and Control Logic, Inc. Apparently, in response to the price competition, DEC has introduced th.e PDP-8/A, incorporating the memory characteristics of the microprocessor-based computers. Relative to older minicomputers, the microprocessor-based systems and the PDP-8/A "miniprocessor" require more time to execute an experimentally useful task. However, their real-time operating characteristics would appear to be sufficient for on-line use in behavioral control and data acquisition. The PDP-8/A's compatibility with older models of the PDP·8 series enables it to use previously developed software, but, for purposes of software development, the microprocessor systems offer a more powerful architecture and instruction set. The introduction of microprocessors would appear to However, the 8-bit word length found in Motorola's offer a relatively inexpensive means for applying 8600 and Intel's 8080 microprocessors, would not place computer technology to the acquisition and control of a serious limitation on the use of microprocessor systems behavioral data. However, as noted by McLean (1974), for control and data acquisition, since a large number of microprocessors, since they are only central processing peripheral devices employ 8-bit codes. Moreover, units (CPUs), require considerable additional circuitry to principally by the use of multiple-word instructions, fulfill memory, timing, and I/O functions to interact 8-bit microprocessors achieve substantial flexibility in with the experimental environment.
    [Show full text]
  • MINICOMPUTER CONCEPTS by BENEDICTO CACHO Bachelor Of
    MINICOMPUTER CONCEPTS By BENEDICTO CACHO II Bachelor of Science Southeastern Oklahoma State University Durant, Oklahoma 1973 Submitted to the Faculty of the Graduate College of the Oklahoma State University in partial fulfillment of the requirements for the Degre~ of MASTER OF SCIENCE July, 1976 . -<~ .i;:·~.~~·:-.~?. '; . .- ·~"' . ~ ' .. .• . ~ . .. ' . ,. , .. J:. MINICOMPUTER CONCEPTS Thesis Approved: fl. F. w. m wd/ I"'' ') 2 I"' e.: 9 tl d I a i. i PREFACE This thesis presents a study of concepts used in the design of minicomputers currently on the market. The material is drawn from research on sixteen minicomputer systems. I would like to thank my major adviser, Dr. Donald D. Fisher, for his advice, guidance, and encouragement, and other committee members, Dr. George E. Hedrick and Dr. James Van Doren, for their suggestions and assistance. Thanks are also due to my typist, Sherry Rodgers, for putting up with my illegible rough draft and the excessive number of figures, and to Dr. Bill Grimes and Dr. Doyle Bostic for prodding me on. Finally, I would like to thank members of my family for seeing me through it a 11 . iii TABLE OF CONTENTS Chapter Page I. INTRODUCTION 1 Objective ....... 1 History of Minicomputers 2 II. ELEMENTS OF MINICOMPUTER DESIGN 6 Introduction 6 The Processor . 8 Organization 8 Operations . 12 The Memory . 20 Input/Output Elements . 21 Device Controllers .. 21 I/0 Operations . 22 III. GENERAL SYSTEM DESIGNS ... 25 Considerations ..... 25 General Processor Designs . 25 Fixed Purpose Register Design 26 General Purpose Register Design 29 Multi-accumulator Design 31 Microprogramm1ng 34 Stack Structures 37 Bus Structures . 39 Typical System Options 41 IV.
    [Show full text]
  • Hardware and Software Considerations for Improving the Throughput of Scientific Computation Computers
    University of Central Florida STARS Retrospective Theses and Dissertations 1977 Hardware and Software Considerations for Improving the Throughput of Scientific Computation Computers Glenn Allen Sullivan University of Central Florida Part of the Engineering Commons Find similar works at: https://stars.library.ucf.edu/rtd University of Central Florida Libraries http://library.ucf.edu This Masters Thesis (Open Access) is brought to you for free and open access by STARS. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of STARS. For more information, please contact [email protected]. STARS Citation Sullivan, Glenn Allen, "Hardware and Software Considerations for Improving the Throughput of Scientific Computation Computers" (1977). Retrospective Theses and Dissertations. 382. https://stars.library.ucf.edu/rtd/382 HARDWARE Al~D SOFTWARE CONSIDERATIONS FOR IMPROVING THE THROUGHPUT OF SCIENTIFIC COMPUTATION -.- COMPUTERS BY GLENN ALLEN SULLIVAN B.S.E. Florida Technological University, 1971 THESIS Submitted in partial fulfillment of the requirements for the degree of Master of Science in Engineering in the Graduate Studies Program of the College of Engineering of Florida Technological University Orlando, Florida 1977 HARDWARE AND SOFTWARE CONSIDERATIONS FOR IMPROVING THE THROUGHPUT OF SCIENTIFIC COMPUTATION COMPurERS by Glenn Allen Sullivan ABSTRAcr In this paper, hardware and software techniques are pre- sented for improving the Throughput (defined as Computations per dollar) of computing systems which are oriented towards high- precision floating point computations. The various improvements are referenced to a baseline of the PDP 11/20, the NOVA 1200, and the TI 960A, all 16 bit minicomputers. The most beneficial hard- ware improvement is the inclusion of a Floating Point Processor, which yields up to 200X Throughput increase over a software floating point package.
    [Show full text]
  • 19800009063.Pdf
    N O T I C E THIS DOCUMENT HAS BEEN REPRODUCED FROM MICROFICHE. ALTHOUGH IT IS RECOGNIZED THAT CERTAIN PORTIONS ARE ILLEGIBLE, IT IS BEING RELEASED IN THE INTEREST OF MAKING AVAILABLE AS MUCH INFORMATION AS POSSIBLE 0 1 MITRE Technical Report MTR•4723 Vol. I TITIS Communications software Volume I- Computer Interfaces J. S. Brown M. D. lenker APRIL 1979 CONTRACT SPONSOR NASA/JSC CONTRACT NO. F19628.79-C-0001 T5295F PROJECT NO. 8470 DEPT. D72 THE MITRE_ This document was prepared for authorized HOUSTON, TEXAS distribution. It has not been approved for public release. T Department Approval; MITRE Project Approval;^''^'^^ a ABSTRACT At NASA's Johnson Space Center MITRE has installed a prototype bus communications system, which is being used to support the Trend Monitoring System (TMS) as well as for evaluation of the bus concept. As a part of the TMS bus installation, MITRE implemented hardware and software interfaces to the MODCOMP and NOVA minicomputers included in the system. This document describes the system software required to drive the interfaces in each TMS computer. Documentation of other software for bus statistics monitoring and for transferring files across the bus is also included. o^ob;Y;Y9e;ti;Y^C^CO'v9e;'ede;4;Yx;Ye8o4;'r;'e;Y^CO4*;t';Y;ti9roY^9e^r^C;k;'cue;4;eat^>Y;koY;'e;'e4eYroY;`e;2;'eoti^v4;Y;Y9e;k;'C;c;'c;4;Y;Y;Y^e^c;'e;Yok;'e^;Yko'e;t NOTICE: THE EQUIPMENT DESCRIBED HEREIN IS THE SUBJECT OF A PATENT APPLICATION PENDING BEFORE THE UNITED STATES PATENT OFFICE.
    [Show full text]