COMPUTATIONAL FLUID DYNAMICS: an Applications Profile from Convex Computer Corporation

Total Page:16

File Type:pdf, Size:1020Kb

COMPUTATIONAL FLUID DYNAMICS: an Applications Profile from Convex Computer Corporation COMPUTATIONAL FLUID DYNAMICS: An Applications Profile from Convex Computer Corporation This simulation. which is based on a Navier-Stokes analysis. depicts surface This simulation shows surface pressure predictions for the AV-88 Harrier II pressure predictions for the F-15 Eagle at Mach 0.9. forebody at Mach 0.8. Supercomputer simulations permit the engineer to observe phenomena that The AV-88 Harrier II can operate from small. unprepared sites and roads close wind tunnels cannot produce. to front Jines. Supercomputing performance brings large-scale simula­ effective resource for innovation, evaluation and enhancement tions into the aircraft designers repertoire. While not expected of aircraft design. to replace such traditional design tools as wind tunnels and test Design development using supercomputing simulation pro­ fiights, computer simulations offer the manufacturer a cost- ceeds from a systematic knowledge base rather than from conjecture or surmise. At the same time, computer simulation investigating the possibilities, MCAIR selected the Convex Cl, the increases the scope for design innovation and enhancement. first affordable supercomputer Its strong price/performance ratio The use of simulations makes iterative design experimentation proved decisive. cost-effective. Moreover, si mulations point the way to improve­ The Cl combines memory and vector processing capabilities ments in design. previously found only on large-scale supercomputers with the Supercomputer simulations also permit software and price advantages of minicom­ the engineer to obseNe details of fluid flow puters. MCAIR found that the Cl completed phenomena that wind tunnels cannot pro­ in 30 minutes runs that required hours on a duce. These include surface pressure dis­ "Computational fluid minicomputer tr i bu ti on s, shock wave locations and dynamics is more complex for Installed in September 1985, this 64-bit strengths, streamline paths and boundary military aircraft than for integrated scalar and vector processor with a layer behavior. In addition, simulations Cray®-like architecture ensured MCAIR a provide critical guidance - before costly transport aircraft and we've superior resource for developing and testing wind tunnel testing - concerning the need found the CJ equal to the code to run on the Cray and location for specific flow diagnostic challenge. Its an ideal The Cl s large physical memory, up to 128 measurements. platform for development of megabytes, provided the capacity to handle Finally, computer simulation has extraor­ code to run on the Cray: it's a larger models. dinary value in training. vector machine, it's open and The Cl at MCAIR is linked to a VAXclus­ The novice engineer who works with it's interactive. The CJ is also ter ™ which in turn is linked to the Cray simulations quickly acquires direct experi­ easy to manage, even in an In day-to-day use, MCAIR has found that ence w ith a broad range of design alterna­ the Cl delivers performance up to 20 times environment such as ours in tives and develops a lasting appreciation of that offered by the VAX TM 780. Moreover, the impact of design changes. which two departments are having supercomputing power on-site The result, for companies that exploit this sharing CPU time:· enables MCAIR to avoid the lengthy turn­ new technology, is superior design. Dr. August Verhoff, around times frequently associated with THE PROBLEM Section Chief, Technology­ time-sharing. In 1984, M cDonnell Aircraft (MCAIR), a The Cl operating system, Convex division of McDonnell Douglas and a lead­ Aerodynamics, McDonnell UNIX®, is enhanced for scientific and engi­ ing manufacturer of state-of-the-art military Aircraft Company neering users. In addition, Convex can sup­ aircraft, undertook a search for a computer "We have to maintain a ply an attractive library o f third-party system that would improve its capabilities for quality development software, math libraries and software tools. development of computational fluid program within stringent The Cl s 110 processors ensure that CPU dynamics code. The primary users of the time is not occupied with user and other 1/0 new resource would be MCAIR~ aero­ budget limitations, and thats A high 1/0 bandwidth to disk and a disk dynamics and propulsion-thermodynamics what makes the CJ such a striping facility enable the Cl to sustain high departments. winner here. We save 110 performance levels. The requirements involved were exacting. tremendously in not having THE ORGANIZATIONS The principal need was a system that would to do time-sharing for code McDonnell Aircraft Corporation manufac­ allow MCAIR to sustain its outstanding development. In fact I get tures high-performance aircraft for military development program while controlling its SS00,000 in value each year customers including the U.S. Navy. the U.S. costs. MCAIR prides itself on its commitment Marine Corps, the U.S. Air Force, the National to investing in the future while achieving from the CJ, and I only Guard, the Royal Air Force, and the govern­ better quality. higher productivity and account for half of its use. ments of Australia, Canada, Israel, Japan, reduced costs. Whats more, we expect the Spain and Saudi Arabia. MCAIR is the home To call a halt to the continuing expense of CJ to be able to handle the of the F-15 fighter, which recently achieved time-sharing on a Cyber ™, the company development load for at least the millionth flight-hour milestone with the resolved to make a sound capital investment another two years:' best safety record of any fighter in the history in a machine that could be located on-site of the U.S. Air Force. Based in St. Louis, Mis­ and fully loaded without incurring user fees. Raymond R. Cosner; souri, McDonnell Aircraft is part of M cDon­ Because work in computational fluid Unit Chief, Propulsion - nell Douglas Corporation. dynamics is CPU-intensive, even in test cases, Thermodynamics Convex Computer Corporation, head­ MCAIR also demanded a machine that was Department McDonnell quartered in Richardson, Texas, was founded top-rated. on a dollars per calculation basis. Aircraft Company in September 1982 to design, manufacture, In addition, since the planned system market and seNice affordable supercompu­ would be used to develop code that would ters. The Convex Cl, which features a Cray­ run on the Cray-1 TM, MCAIR needed a power­ like architecture, combines 64-bit scalar and ful vector processing machine. vector processing capabilities of large-scale supercomputers THE SOLUTION with software and price advantages of minicomputers. Offices These requirements considerably narrowed the range of alter­ worldwide ensure service and support to customers using natives. Their high costs quickly put Cyber machines out of the Convex systems for geophysical research, computer-aided engi­ running. Their slower speeds ruled out minicomputers. After neering and design, defense, aerospace and general research. Convex and the Convex logo are trademarks of Convex Computer Corporation. Cyber is a trademark of Control Data Corporation. Cray is a registered trademark and Cray-I is a trademark of Cray Research. Inc. VAX and VAXcluster are trademarks of Digital Equipment Corporation. UNIX is a registered trademark of AT&T Although the material contained.herein has been carefully revievved, Convex Computer Corporation (Convex) does not warrant it to be free of errors or omissions. Convex reserves the right to make corrections. updates. rev1s1ons or changes to the information contained herein. Convex does not warrant the material described herein to be free of patent infringement. Convex Computer Corporation. 701 North Plano Road. Richardson. Texas 75081 (214) 952-0200 Publication 080-002187-000 11/86 Printed in the U.S.A. .
Recommended publications
  • Emerging Technologies Multi/Parallel Processing
    Emerging Technologies Multi/Parallel Processing Mary C. Kulas New Computing Structures Strategic Relations Group December 1987 For Internal Use Only Copyright @ 1987 by Digital Equipment Corporation. Printed in U.S.A. The information contained herein is confidential and proprietary. It is the property of Digital Equipment Corporation and shall not be reproduced or' copied in whole or in part without written permission. This is an unpublished work protected under the Federal copyright laws. The following are trademarks of Digital Equipment Corporation, Maynard, MA 01754. DECpage LN03 This report was produced by Educational Services with DECpage and the LN03 laser printer. Contents Acknowledgments. 1 Abstract. .. 3 Executive Summary. .. 5 I. Analysis . .. 7 A. The Players . .. 9 1. Number and Status . .. 9 2. Funding. .. 10 3. Strategic Alliances. .. 11 4. Sales. .. 13 a. Revenue/Units Installed . .. 13 h. European Sales. .. 14 B. The Product. .. 15 1. CPUs. .. 15 2. Chip . .. 15 3. Bus. .. 15 4. Vector Processing . .. 16 5. Operating System . .. 16 6. Languages. .. 17 7. Third-Party Applications . .. 18 8. Pricing. .. 18 C. ~BM and Other Major Computer Companies. .. 19 D. Why Success? Why Failure? . .. 21 E. Future Directions. .. 25 II. Company/Product Profiles. .. 27 A. Multi/Parallel Processors . .. 29 1. Alliant . .. 31 2. Astronautics. .. 35 3. Concurrent . .. 37 4. Cydrome. .. 41 5. Eastman Kodak. .. 45 6. Elxsi . .. 47 Contents iii 7. Encore ............... 51 8. Flexible . ... 55 9. Floating Point Systems - M64line ................... 59 10. International Parallel ........................... 61 11. Loral .................................... 63 12. Masscomp ................................. 65 13. Meiko .................................... 67 14. Multiflow. ~ ................................ 69 15. Sequent................................... 71 B. Massively Parallel . 75 1. Ametek.................................... 77 2. Bolt Beranek & Newman Advanced Computers ...........
    [Show full text]
  • The Risks Digest Index to Volume 11
    The Risks Digest Index to Volume 11 Search RISKS using swish-e Forum on Risks to the Public in Computers and Related Systems ACM Committee on Computers and Public Policy, Peter G. Neumann, moderator Index to Volume 11 Sunday, 30 June 1991 Issue 01 (4 February 1991) Re: Enterprising Vending Machines (Allan Meers) Re: Risks of automatic flight (Henry Spencer) Re: Voting by Phone & public-key cryptography (Evan Ravitz) Re: Random Voting IDs and Bogus Votes (Vote by Phone) (Mike Beede)) Re: Patriots ... (Steve Mitchell, Steven Philipson, Michael H. Riddle, Clifford Johnson) Re: Man-in-the-loop on SDI (Henry Spencer) Re: Broadcast local area networks ... (Curt Sampson, Donald Lindsay, John Stanley, Jerry Leichter) Issue 02 (5 February 1991) Bogus draft notices are computer generated (Jonathan Rice) People working at home on important tasks (Mike Albaugh) Predicting system reliability (Martyn Thomas) Re: Patriots (Steven Markus Woodcock, Mark Levison) Hungry copiers (another run-in with technology) (Scott Wilson) Re: Enterprising Vending Machines (Dave Curry) Broadcast LANs (Peter da Silva, Scott Hinckley) Issue 03 (6 February 1991) Tube Tragedy (Pete Mellor) New Zealand Computer Error Holds Up Funds (Gligor Tashkovich) "Inquiry into cash machine fraud" (Stella Page) Quick n' easy access to Fidelity account info (Carol Springs) Re: Enterprising Vending Machines (Mark Jackson) RISKS of no escape paths (Geoff Kuenning) A risky gas pump (Bob Grumbine) Electronic traffic signs endanger motorists... (Rich Snider) Re: Predicting system reliability (Richard P. Taylor) The new California licenses (Chris Hibbert) Phone Voting -- Really a Problem? (Michael Barnett, Dave Smith) Re: Electronic cash completely replacing cash (Barry Wright) Issue 04 (7 February 1991) Subway door accidents (Mark Brader) http://catless.ncl.ac.uk/Risks/index.11.html[2011-06-11 08:17:52] The Risks Digest Index to Volume 11 "Virus" destroys part of Mass.
    [Show full text]
  • Some Performance Comparisons for a Fluid Dynamics Code
    NBSIR 87-3638 Some Performance Comparisons for a Fluid Dynamics Code Daniel W. Lozier Ronald G. Rehm U.S. DEPARTMENT OF COMMERCE National Bureau of Standards National Engineering Laboratory Center for Applied Mathematics Gaithersburg, MD 20899 September 1987 U.S. DEPARTMENT OF COMMERCE NATIONAL BUREAU OF STANDARDS NBSIR 87-3638 SOME PERFORMANCE COMPARISONS FOR A FLUID DYNAMICS CODE Daniel W. Lozier Ronald G. Rehm U.S. DEPARTMENT OF COMMERCE National Bureau of Standards National Engineering Laboratory Center for Applied Mathematics Gaithersburg, MD 20899 September 1987 U.S. DEPARTMENT OF COMMERCE, Clarence J. Brown, Acting Secretary NATIONAL BUREAU OF STANDARDS, Ernest Ambler, Director - 2 - 2. BENCHMARK PROBLEM In this section we describe briefly the source of the benchmark problem, the major logical structure of the Fortran program, and the parameters of three different specific instances of the benchmark problem that vary widely in the amount of time and memory required for execution. Research Background As stated in the introduction, our purpose in benchmarking computers is solely in the interest of further- ing our investigations into fundamental problems of fire science. Over a decade ago, stimulated by federal recognition of very large losses of fife and property by fires each year throughout the nation, NBS became actively involved in a national effort to reduce such losses. The work at NBS ranges from very practical to quite theoretical; our approach, which proceeds directly from basic principles, is at the theoretical end of this spectrum. Early work was concentrated on developing a mathematical model of convection arising from a prescribed source of heat in an enclosure, e.g.
    [Show full text]
  • Programming Languages & Tools
    I PROGRAMMING LANGUAGES & TOOLS Volume 10 Number 1 1998 Editorial The Digital Technicaljoumalis a refereed AlphaServer, Compaq, tl1e Compaq logo, jane C. Blake, Managing Editor journal published quarterly by Compaq DEC, DIGITAL, tl1e DIGITAL logo, 550 ULTIUX, Kathleen M. Stetson, Editor Computer Corporation, King Street, VAX,and VMS are registered 01460-1289. Hden L. Patterson, Editor LKGI-2jW7, Littleton, MA in the U.S. Patent and Trademark Office. Hard-copy subscriptions can be ordered by DIGITAL UNIX, FX132, and OpenVMS Circulation sending a check in U.S. funds (made payable arc trademarks of Compaq Computer Kristine M. Lowe, Administrator to Compaq Computer Corporation) to the Corporation. published-by address. General subscription Production rates arc $40.00 (non-U.S. $60) for four issues Intel and Pentium are registered u·ademarks $75.00 $115) Christa W. Jessica, Production Editor and (non-U.S. for eight issues. of Intel Corporation. University and college professors and Ph.D. Elizabeth McGrail, Typographer I lUX is a registered trademark of Silicon students in the elecu·icaJ engineering and com­ Peter R. Woodbury, Illustrator Graphics, Inc. puter science fields receive complimentary sub­ scriptions upon request. Compaq customers Microsoft, Visual C++, Windows, and Advisory Board may qualify tor giftsubscriptions and arc encour­ Windows NT are registered trademarks Thomas F. Gannon, Chairman (Acting) aged to contact tl1eir sales representatives. of Microsoft Corporation. Scott E. Cutler Donald Z. Harbert Electronic subscriptions are available at MIPS is a registered trademark of MIPS William A. Laing no charge by accessing URL Technologies, Inc. Richard F. Lary http:jjwww.digital.com/subscription.
    [Show full text]
  • Certification and Benchmarking of AIPS on the Convex
    AIPS MEMO NO. _ 3 £ Certifioation and Benchmarking of AIPS on the Convex C-l and Alliant FX/8 Kerry C. Hilldrup, Donald C. Wells, William D. Cotton National Radio Astronomy Observatory [1] Edgemont Road, Charlottesville, VA 22903-2475 (804)296-0211, FTS=938-1271, TWX=910-997-0174 24 December 1985 [2] Abstract The 15APR85 release of AIPS has been installed on the Convex C-l vector computer and on the Alliant FX/8 vector/concurrent computer. Both have been certified using the PFT benchmarking and certification test. Although a small number of compiler bugs were encountered in each, the AIPS application code was installed with only minimal modifications, and computed results agreed well with other implementations of AIPS. In the course of the implementations, the technology of the Clark CLEAN algorithm was advanced considerably; the final vectorized CLEAN algorithm on both systems is about three times as fast as the current microcode algorithm on the FPS AP-120B array processor. Similar improvements were observed for other highly vectorized tasks. Programs which were not vectorized generally executed on both in comparable CPU times and faster real times than they achieved on the DEC VAX-8600. The FX/8 with 6 computational elements (CEs) generally outperformed the C-l by a little in CPU time, but was significantly slower in real time. The performance of the FX/8 as a function of the number of CEs is also described. [1] NRAO is operated by Associated Universities, Inc., under contract with the U. S. National Science Foundation. [2] this report entirely supersedes and replaces a previous version which was dated 18 July 1985 and was marked as "AIPS Memo No.
    [Show full text]
  • Supercomputers: the Amazing Race Gordon Bell November 2014
    Supercomputers: The Amazing Race Gordon Bell November 2014 Technical Report MSR-TR-2015-2 Gordon Bell, Researcher Emeritus Microsoft Research, Microsoft Corporation 555 California, 94104 San Francisco, CA Version 1.0 January 2015 1 Submitted to STARS IEEE Global History Network Supercomputers: The Amazing Race Timeline (The top 20 significant events. Constrained for Draft IEEE STARS Article) 1. 1957 Fortran introduced for scientific and technical computing 2. 1960 Univac LARC, IBM Stretch, and Manchester Atlas finish 1956 race to build largest “conceivable” computers 3. 1964 Beginning of Cray Era with CDC 6600 (.48 MFlops) functional parallel units. “No more small computers” –S R Cray. “First super”-G. A. Michael 4. 1964 IBM System/360 announcement. One architecture for commercial & technical use. 5. 1965 Amdahl’s Law defines the difficulty of increasing parallel processing performance based on the fraction of a program that has to be run sequentially. 6. 1976 Cray 1 Vector Processor (26 MF ) Vector data. Sid Karin: “1st Super was the Cray 1” 7. 1982 Caltech Cosmic Cube (4 node, 64 node in 1983) Cray 1 cost performance x 50. 8. 1983-93 Billion dollar SCI--Strategic Computing Initiative of DARPA IPTO response to Japanese Fifth Gen. 1990 redirected to supercomputing after failure to achieve AI goals 9. 1982 Cray XMP (1 GF) Cray shared memory vector multiprocessor 10. 1984 NSF Establishes Office of Scientific Computing in response to scientists demand and to counteract the use of VAXen as personal supercomputers 11. 1987 nCUBE (1K computers) achieves 400-600 speedup, Sandia winning first Bell Prize, stimulated Gustafson’s Law of Scalable Speed-Up, Amdahl’s Law Corollary 12.
    [Show full text]
  • Shared Memory Programming on NUMA–Based Clusters Using a General and Open Hybrid Hardware / Software Approach
    Shared Memory Programming on NUMA–based Clusters using a General and Open Hybrid Hardware / Software Approach Martin Schulz Institut für Informatik Lehrstuhl für Rechnertechnik und Rechnerorganisation Shared Memory Programming on NUMA–based Clusters using a General and Open Hybrid Hardware / Software Approach Martin Schulz Vollständiger Abdruck der von der Fakultät für Informatik der Technischen Universität München zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften (Dr. rer. nat.) genehmigten Dissertation. Vorsitzender: Univ.-Prof. R. Bayer, Ph.D. Prüfer der Dissertation: 1. Univ.-Prof. Dr. A. Bode 2. Univ.-Prof. Dr. H. Hellwagner, Universität Klagenfurt / Österreich Die Dissertation wurde am 24. April 2001 bei der Technischen Universität München ein- gereicht und durch die Fakultät für Informatik am 28. Juni 2001 angenommen. Abstract The widespread use of shared memory programming for High Performance Computing (HPC) is currently hindered by two main factors: the limited scalability of architectures with hardware support for shared memory and the abundance of existing programming models. In order to solve these issues, a comprehensive shared memory framework needs to be created which enables the use of shared memory on top of more scalable architectures and which provides a user–friendly solution to deal with the various different programming models. Driven by the first issue, a large number of so–called SoftWare Distributed Shared Memory (SW–DSM) systems have been developed. These systems rely solely on soft- ware components to create a transparent global virtual memory abstraction on highly scal- able, loosely coupled architectures without any direct hardware support for shared memory. However, they are often affected by inherent performance problems and, in addition, do not solve the second issue of the existence of (too) many shared memory programming models.
    [Show full text]
  • Shared Memory Multiprocessor-A Brief Study
    © 2014 IJIRT | Volume 1 Issue 6 | ISSN : 2349-6002 SHARED MEMORY MULTIPROCESSOR-A BRIEF STUDY Shivangi Kukreja,Nikita Pahuja,Rashmi Dewan Student,Computer Science&Engineering,Maharshi Dayanand University Gurgaon,Haryana,India Abstract- The intention of this paper is to provide an The popularity of these systems is not due simply to overview on the subject of Advanced computer the demand for high performance computing. These Architecture. The overview includes its presence in systems are excellent at providing high throughput hardware and software and different types of for a multiprocessing load, and function effectively organization. This paper also covers different part of as high-performance database servers, network shared memory multiprocessor. Through this paper we are creating awareness among the people about this servers, and Internet servers. Within limits, their rising field of multiprocessor. This paper also offers a throughput is increased linearly as more processors comprehensive number of references for each concept are added. in SHARED MEMORY MULTIPROCESSOR. In this paper we are not so interested in the performance of database or Internet servers. That is I. INTRODUCTION too passé; buy more processors, get better In the mid-1980s, shared-memory multiprocessors throughput. We are interested in pure, raw, were pretty expensive and pretty rare. Now, as unadulterated compute speed for our high hardware costs are dropping, they are becoming performance application. Instead of running hundreds commonplace. Many home computer systems in the of small jobs, we want to utilize all $750,000 worth under-$3000 range have a socket for a second CPU. of hardware for our single job.
    [Show full text]
  • HP Employees Have the Responsibility for Keeping the HP Received a Good Deal Organization Chart up to Date
    FROM THE EDITOR ithout even knowing And along the way, Betty took on it, HP employees have the responsibility for keeping the HP received a good deal organization chart up to date. She oftheir daily HP news meticulously has tracked every revi­ for the past 20 years sion, reassignment, realignment and from one person. reorganization. IfWalter Cronkite is considered Considering how inaccurate a the most trusted person in American printed org chart can be, the one on journalism, then the HP equivalent is Betty's office wall-marked up, Betty Gerard. crossed over, penciled-in and dotted For two decades, Betty has been with yellow Post-it squares-has been HP's most important news source. the only truly accurate org chart Ask Betty the correct name of an HP anywhere in HP for most of the past On the cover: Using HP division and she probably can tell you several years. analytical equipment, Dr. its complete history-when the divi­ Betty is a stickler for detail and one Holmes Morton diagnosed and now successfully treats sion began, every name it's had, a of the few people I know who gets Amish children who suffer chronological list of its general man­ downright passionate about conunas. from a hereditary disease. agers and probably a list of every We've been known to debate the mer­ The photo feature begins on page 4. Cover photo by product manufactured there. its of what she calls a "gratuitous Clark Mishler. If Betty doesn't know the answer comma" for days on end. from memory, it's sure to reside on a And now, after 20 years with HP, neatly typed 3-by-5-inch index card in Betty has retired.
    [Show full text]
  • A Publication of Hewlett-Packard November 1992
    A Publication of Hewlett-Packard A A November 1992 The new HP industrial workstations and real-time computer systems combine the performance, support, and service that form the foundation of HP's reputation. seepage12 Programisdesignedto help more users apply the power of multiple workstations to large, numerically intensive applications. , Page 1 i The five new HP JetDirect cards for Ethernet and AppleTalk networks support and enablc users to connect perivhertrl devices black and color print- ing using both the blac!k cmd color cartridges together. seepage2 HPVol. 2, No. 10 ComuterNovember 1992 A Update I HP Computer Update is a monthly publication To order HP software or hardware products, call of Hewlett-Pxkard Company that provides your HP representative. A telephone listing for infodonon new Hewlett-Packard computer sales offices in the U.S. and Canada appears on the software and hardware products, events, and back cover. promotions. To receive literature or technical information on Mmaghg Editor: Traey Wester HP computer systems and worhtations, and to order workstations and select computer products, Address editorial correspondence and address call HP DIRECT at 800-637-7740 (U.S. only). coJ.Td0ns to: To receive technical information about any HP HP Cornpter Update product, call the Customer Information Center: P.O.Box 51827 800-752-0900 in the U.S., or 8003873867 in Palo Alto, CA 943030724 Canada Distribution information I hY Computer Update is distributed to individuals who are involved in spewor purchasing HP computer products and systems. It is also sent to HP worldwide value-added business partners. If you would like to receive a copy, please contact your local HP representative.
    [Show full text]
  • Independent PA-RISC and Itanium Reference Book
    OpenPA The book of PA-RISC Paul Weissmann Bonn This document and its content are Copyright © 1999-2021 Paul Weissmann, Bonn, Germany, unless otherwise noted. Berlin, Bonn, Palo Alto, San Francisco No parts of this document may be reproduced or copied without prior written permission. Commercial use of the content is prohibited. OpenPA.net, the information resource on PA-RISC. Second edition Release 2.8 Bonn, May 2021 Other editions: Second edition 2.7: July 2020 Second edition 2.6: January 2018 Second edition 2.5: January 2016 Second edition 2.4: July 2012 Second edition 2.3: July 2009 Second edition 2.2: January 2009 Second edition 2.1: October 2008 Second edition 2.0: May 2008 First edition 1.2: December 2007 First edition 1.1: November 2007 First edition 1.0: July 2006 OpenPA.net (online) is a registered serial publication, ISSN 1866-2757. Paul Weissmann can be reached by e-mail: [email protected]. and online at Insignals Cyber Security and OpenKRITIS. Preface This is the print edition of the OpenPA.net website from Spring 2021. OpenPA is a resource for HP PA-RISC and Itanium computers with technical descriptions of workstations, servers, their hardware architecture and supported operating systems This project is independent of and does not represent The Hewlett Packard Company in any way. This is the Second Edition 2.8. Set with LATEX. Changes in Second Edition 2.8 since the last edition (2020): ê Many revisions and corrections (thanks!) ê Text and language in all chapters ê TeX backend update All other changes are listed in chapter 5.1.
    [Show full text]
  • A Study of Workstation Computational Performance for Real-Time Flight Simulation
    / NASA Technical Memorandum 109184 A Study of Workstation Computational Performance for Real-Time Flight Simulation Jeffrey M. Maddalon and Jeff I. Cleveland II Langley Research Center, Hampton, Virginia (NASA-TM-lO?I84) A STUOY OF WORKSTATION COMPUTATIONAL N95-2724I PERFORMANCE FOR REAL-TIME FLIGHT SIMULATION (NASA. lang|ey Research Center) ]0 p Unclas G3/60 0049832 February 1995 National Aeronautics and Space Administration Langley Research Center Hampton, Virginia 23681-0001 A Study of Workstation Computational Performance for Real-Time Flight Simulation Jeffrey M. Maddalon JeffI. Cleveland II Summary This paper presents the results of a computational benchmark, based on actual real-time flight simulation code of an X-29 aircraft used at Langley Research Center. This benchmark was run on workstations from Digital Equipment Corporation, Hewlett-Packard, International Business Machines, Silicon Graphics, and Sun Microsystems. The intent of this study is to measure the computational suitability of workstations to operate a simulation model of an X-29 aircraft. Before computational performance can be considered relevant, computational accuracy and software porting costs must be found to be acceptable. This study indicates that, in general, workstation vendors have no problem meeting computational accuracy requirements for the X-29 aircraft simulation model. Porting this simulation model to different computational platforms shows that there is little middle ground, porting will be either easy or difficult. The computational performance results show that workstations from several vendors can provide the necessary computational power (along with sufficient computational accuracy and moderate software porting costs) to properly operate a real-time flight simulation model of an X-29 aircraft.
    [Show full text]