Overview and History Nas Overview
Total Page:16
File Type:pdf, Size:1020Kb
Load more
Recommended publications
-
A Look at the Impact of High-End Computing Technologies on NASA Missions
A Look at the Impact of High-End Computing Technologies on NASA Missions Contributing Authors Rupak Biswas Jill Dunbar John Hardman F. Ron Bailey Lorien Wheeler Stuart Rogers Abstract From its bold start nearly 30 years ago and continuing today, the NASA Advanced Supercomputing (NAS) facility at Ames Research Center has enabled remarkable breakthroughs in the space agency’s science and engineering missions. Throughout this time, NAS experts have influenced the state-of-the-art in high-performance computing (HPC) and related technologies such as scientific visualization, system benchmarking, batch scheduling, and grid environments. We highlight the pioneering achievements and innovations originating from and made possible by NAS resources and know-how, from early supercomputing environment design and software development, to long-term simulation and analyses critical to design safe Space Shuttle operations and associated spinoff technologies, to the highly successful Kepler Mission’s discovery of new planets now capturing the world’s imagination. Keywords Information Technology and Systems; Aerospace; History of Computing; Parallel processors, Processor Architectures; Simulation, Modeling, and Visualization The story of NASA’s IT achievements is not complete without a chapter on the important role of high-performance computing (HPC) and related technologies. In particular, the NASA Advanced Supercomputing (NAS) facility—the space agency’s first and largest HPC center—has enabled remarkable breakthroughs in NASA’s science and engineering missions, from designing safer air- and spacecraft to the recent discovery of new planets. Since its bold start nearly 30 years ago and continuing today, NAS has led the way in influencing the state-of-the-art in HPC and related technologies such as high-speed networking, scientific visualization, system benchmarking, batch scheduling, and grid environments. -
Paper in PDF Format
Managing System Resources on ‘Teras’ Experiences in Managing System Resources on a Large SGI Origin 3800 Cluster The paper to the presentation given at the CUG SUMMIT 2002 held at the University of Manchester. Manchester (UK), May 20-24 2002. http://www.sara.nl Mark van de Sanden [email protected] Huub Stoffers [email protected] Introduction Outline of the Paper This is a paper about our experiences with managing system resources on ‘Teras’. Since November 2000 Teras, a large SGI Origin 3800 cluster, is the national supercomputer for the Dutch academic community. Teras is located at and maintained by SARA High Performance Computing (HPC). The outline of this presentation is as follows: Very briefly, we will tell you what SARA is, and what we presently do, particularly in high performance computing, besides maintaining Teras. Subsequently, to give you an idea, on the one hand which resources are available on Teras, and on the other hand what resources are demanded, we will give overviews of three particular aspects of the Teras: 1) Some data on the hardware – number of CPUs, memory, etc. - and the cluster configuration, i.e. description of the hosts that constitute the cluster. 2) Identification of the key software components that are used on the system; both, system software components, as well as software toolkits available to users to create their (parallel) applications. 3) A characterization of the ‘job mix’ that runs on the Teras cluster. Thus equipped with a clearer idea of both ‘supply’ and ‘demand’ of resources on Teras, we then state the resource allocation policy goals that we pursue in this context, and we review what resource management possibilities we have at hand and what they can accomplish for us, in principle. -
Ribbons Manual Help Ribbons On-Line Manual
Ribbons Manual Help Ribbons On-Line Manual. Usage: ribbons [options] ribbons is driven by mouse and an X/Motif interface, once data has been prepared with auxiliary programs. Coordinate files must be in Protein Data Bank (*.pdb) format. Recommend that you make a new directory and copy over a *.pdb file to begin. To initialize for default drawing data from a PDB entry, type: ribbons -e YourMol.pdb To display once the data above is prepared, type: ribbons -n YourMol To create any data from 'YourMol.pdb' with ribbons-data GUI, type: ribbons-data YourMol The choices from the "Help" submenus bring up the additional information screens listed below: Data Preparation ● Required Files ● Auxiliary Programs ● Tutorial Overview ● X-ray Error Analysis Graphics ● Motif MenuBar ● Control Panels ● Model Transformation ● Menubar Options ❍ File ❍ Edit ❍ Model ❍ View ❍ Help ● The Ribbon Drawing Panels ❍ Ribbon Style file:///C|/My Documents/ribbons/help/manual.html (1 of 2) [4/3/2001 1:52:00 PM] Ribbons Manual Help ❍ Ribbon Dimensions ❍ Ribbon N-primitives ❍ Ribbon Palette ● Primitive Control Panels ❍ Atom Panel ❍ Bond Panel ❍ Poly Panel ❍ Ndot Panel ❍ Text Panel ● Material and Image Control Panels ❍ Image Panel ❍ Color Panel ❍ Light Panel ❍ colorIndex Panel ❍ Motion Panel ● Keyboard Shortcuts Images ● Saving Images ● Example Images ● Artistic Images ● Color Indexing Scheme ● PostScript Plots ● Raytracing ● Virtual Reality Program ● Frequently Asked Questions ● Licensing ● Source Code ● Ribbon References Ribbons User Manual / UAB-CBSE / [email protected] file:///C|/My Documents/ribbons/help/manual.html (2 of 2) [4/3/2001 1:52:00 PM] Ribbons Data Preparation Help Ribbons Data Preparation Ribbons requires generation of a database of files for display. -
Fast Synchronization on Shared-Memory Multiprocessors: an Architectural Approach
Fast Synchronization on Shared-Memory Multiprocessors: An Architectural Approach Zhen Fang1, Lixin Zhang2, John B. Carter1, Liqun Cheng1, Michael Parker3 1 School of Computing University of Utah Salt Lake City, UT 84112, U.S.A. fzfang, retrac, [email protected] Telephone: 801.587.7910 Fax: 801.581.5843 2 IBM Austin Research Lab 3 Cray, Inc. 11400 Burnet Road, MS 904/6C019 1050 Lowater Road Austin, TX 78758, U.S.A. Chippewa Falls, WI 54729, U.S.A. [email protected] [email protected] Abstract Synchronization is a crucial operation in many parallel applications. Conventional synchronization mechanisms are failing to keep up with the increasing demand for efficient synchronization operations as systems grow larger and network latency increases. The contributions of this paper are threefold. First, we revisit some representative synchronization algorithms in light of recent architecture innovations and provide an example of how the simplifying assumptions made by typical analytical models of synchronization mechanisms can lead to significant performance estimate errors. Second, we present an architectural innovation called active memory that enables very fast atomic operations in a shared-memory multiprocessor. Third, we use execution-driven simulation to quantitatively compare the performance of a variety of synchronization mechanisms based on both existing hardware techniques and active memory operations. To the best of our knowledge, synchronization based on active memory outforms all existing spinlock and non-hardwired barrier implementations by a large margin. Keywords: distributed shared-memory, coherence protocol, synchronization, barrier, spinlock, memory controller This effort was supported by the National Security Agency (NSA), the Defense Advanced Research Projects Agency (DARPA) and Silicon Graphics Inc. -
Remanufactured SGI™ Origin™ 2000 High-Performance Server
Product Guide Remanufactured SGI™ Origin™ 2000 High-Performance Server Origin 2000 Rack Multirack System The rack-mounted Connect multiple racks as a shared- Origin 2000 system memory powerhouse or redeploy is the industry’s those assets into multiple, tightly most powerful and integrated departmental servers. flexible shared-memory Add IRIS FailSafe™ software to server. Easily expand provide failover between systems, the system by adding and Origin 2000 becomes the additional racks. ultimate high-availability server for critical computational and business solutions. Origin 2000 Deskside Revolutionary Server for Your Expanding Business SGI server family, can expand from dual-processor With its outstanding Today’s corporate environment is characterized by deskside systems to powerful 512-processor scalable price/performance, the a rapid growth in communications requirements servers without disruptive and expensive box-swaps deskside Origin 2000 and continual organizational change. A great deal common with other server solutions. System server is ideal for of energy is spent planning future system purchases resources can be deployed as tightly integrated evolving applications in an ongoing effort to keep system capabilities arrays of small Origin 2000 systems or as single requiring expansion capability. Seamless upgradability to larger ahead of application demands. Corporations require large shared-memory systems capable of meeting rack-based configurations supports scalable, flexible computer systems that can adapt the most demanding data processing and your growing requirements. to business change and satisfy growing server computing requirements. requirements. SGI offers you a proven, more adaptable way to purchase your data processing, SGI pioneered the development and delivery of computation, and communications capabilities. shared-memory parallel processing systems. -
Infiniband Shines on the Top500 List of World's Fastest Computers
InfiniBand Shines on the Top500 List of World’s Fastest Computers InfiniBand Drives Two Top Ten Supercomputers and Gains Widespread Adoption for Clustering and Storage PITTSBURGH, PA - November 9, 2004 - (SuperComputing 2004) – Mellanox® Technologies Ltd, the leader in performance business and technical computing interconnects, announced today that the InfiniBand interconnect is used on more than a dozen systems on the prestigious Top500 list of the world's fastest computers, including two systems which achieved a top ten ranking. The latest Top500 list (www.top500.org) was unveiled this week at SuperComputing 2004 – the world’s leading conference on high performance computing (HPC). This represents more than a 400% growth since last year when InfiniBand technology made its debut on the list with just three entries. The Top500 computer list is a bellwether technology indicator for the broader performance busi- ness computing and technical computing markets. InfiniBand is being broadly adopted for numer- ically intensive applications including CAD/CAM, oil and gas exploration, fluid dynamics, weather modeling, molecular dynamics, gene sequencing, etc. These applications benefit from the high throughput, low latency InfiniBand interconnect - which allows industry standard servers to be connected together to create powerful computers at unmatched price/performance levels. These InfiniBand clustered computers deliver a 20X price/performance advantage vs. traditional mainframe systems. InfiniBand enables NASA’s Columbia supercomputer to claim its status as the second fastest computer in the world. Columbia uses InfiniBand to connect twenty Itanium-based Silicon Graphics platforms into a 10,240 processor supercomputer that achieves over 51.8 Tflops of sus- Mellanox Technologies Inc. -
IRIX® Device Driver Programmer's Guide
IRIX® Device Driver Programmer’s Guide Document Number 007-0911-210 CONTRIBUTORS Written by David Cortesi, John Raithel, Bill Tuthill, and Anita Manders UpdatedbyJulieBoneyandStevenLevine Illustrated by Dany Galgani, Cheri Brown, and Chrystie Danzer Production by Karen Jacobson COPYRIGHT © 1998-2003 Silicon Graphics, Inc. All rights reserved; provided portions may be copyright in third parties, as indicated elsewhere herein. No permission is granted to copy, distribute, or create derivative works from the contents of this electronic documentation in any manner, in whole or in part, without the prior written permission of Silicon Graphics, Inc. LIMITED RIGHTS LEGEND The electronic (software) version of this document was developed at private expense; if acquired under an agreement with the USA government or any contractor thereto, it is acquired as "commercial computer software" subject to the provisions of its applicable license agreement, as specified in (a) 48 CFR 12.212 of the FAR; or, if acquired for Department of Defense units, (b) 48 CFR 227-7202 of the DoD FAR Supplement; or sections succeeding thereto. Contractor/manufacturer is Silicon Graphics, Inc., 1600 Amphitheatre Pkwy 2E, Mountain View, CA 94043-1351. TRADEMARKS AND ATTRIBUTIONS Silicon Graphics, SGI, the SGI logo, Challenge, Indigo, IRIS, IRIX, O2, Octane, Onyx, Onyx2, and Origin are registered trademarks and Indigo2,Indigo2 Maximum Impact, IRIS InSight, Power Challenge, Power Channel, Power Indigo2,PowerOnyx,andREACT/proare trademarks of Silicon Graphics, Inc., in the United States and/or other countries worldwide. Indy is a registered trademark, used under license in the United States and owned by Silicon Graphics,Inc.inothercountriesworldwide. IBM is a trademark of International Business Machines Corporation. -
William Thigpen Deputy Project Manager August 15, 2011 Introduction
The High End Computing Capability Project: Excellence at the Leading Edge of Computing William Thigpen Deputy Project Manager August 15, 2011 Introduction • NAS has stellar history of amazing accomplishments, drawing from synergy between high end computing (HEC) and modeling and simulation (M&S) • NAS has earned national and international reputation as a pioneer in the development and application of HEC technologies, providing its diverse customers with world-class M&S expertise, and state-of-the-art supercomputing products and services • HECC has been a critical enabler for mission success throughout NASA – it supports the modeling, simulation, analysis, and decision support activities for all four Mission Directorates, NESC, OCE, and OS&MA • HECC Mission: Develop and deliver the most productive HEC environment in the world, enabling NASA to extend technology, expand knowledge, and explore the universe • SCAP model of baseline funding + marginal investment has been immensely successful and has had a positive impact on NASA science and engineering – Baseline funding provides “free” HECC resources and services (MDs select own projects, determine priorities, and distribute own resources) – Marginal investments from ARMD, ESMD, and SMD demonstrate current model has strong support and buy-in from MDs HEC Characteristics • HEC, by its very nature, is required to be at the leading edge of computing technologies – Unlike agency-wide IT infrastructure, HEC is fundamentally driven to provide a rapidly increasing capability to a relatively small -
SGI™ Origin 3000 Series Technical Configuration Owner's Guide
SGI™ Origin 3000 Series Technical Configuration Owner’s Guide Document Number 007-4311-002 CONTRIBUTORS Written by Dick Brownell Illustrated by Dan Young Production by Diane Ciardelli Engineering contributions by SN1 Electrical, Mechanical, Manufacturing, Marketing, and Site Planning teams COPYRIGHT © 2000 Silicon Graphics, Inc. All rights reserved; provided portions may be copyright in third parties, as indicated elsewhere herein. No permission is granted to copy, distribute, or create derivative works from the contents of this electronic documentation. The contents of this document may not be copied or duplicated in any form, in whole or in part, without the prior written permission of Silicon Graphics, Inc. LIMITED RIGHTS LEGEND The electronic (software) version of this document was developed at private expense; if acquired under an agreement with the USA government or any contractor thereto, it is acquired as “commercial computer software” subject to the provisions of its applicable license agreement, as specified in (a) 48 CFR 12.212 of the FAR; or if acquired for Department of Defense units, (b) 48 CFR 227-7202 of the DoD FAR Supplement; or sections succeeding thereto. Contractor/manufacturer is Silicon Graphics, Inc., 1600 Amphitheatre Pkwy. 2E, Mountain View, CA 94043-1351. TRADEMARKS AND ATTRIBUTIONS Silicon Graphics is a registered trademark and SGI, the SGI logo, Origin, Onyx3, and IRIS InSight are trademarks of Silicon Graphics, Inc. PostScript is a trademark of Adobe Systems, Inc. MIPS is a registered trademark of MIPS Technologies, Inc. Windows and Windows NT are registered trademarks of Microsoft Corporation. DVCPRO is a trademark of Panasonic, Inc. Linux is a registered trademark of Linus Torvalds. -
Jack 5 User's Guide
University of Pennsylvania ScholarlyCommons Technical Reports (CIS) Department of Computer & Information Science October 1991 Jack 5 User's Guide Cary B. Phillips University of Pennsylvania Follow this and additional works at: https://repository.upenn.edu/cis_reports Recommended Citation Cary B. Phillips, "Jack 5 User's Guide", . October 1991. University of Pennsylvania Department of Computer and Information Science Technical Report No. MS-CIS-91-78. This paper is posted at ScholarlyCommons. https://repository.upenn.edu/cis_reports/479 For more information, please contact [email protected]. Jack 5 User's Guide Abstract This chapter is a brief tutorial introduction to using Jack. It demonstrates some of the basic Jack features, illustrating mostly the flavor of how you interact with Jack rather than describing the details of how to use it. The first time ouy run Jack, you should go through the examples in this chapter. From then on, refer to the later chapters for more details about the various features. Comments University of Pennsylvania Department of Computer and Information Science Technical Report No. MS- CIS-91-78. This technical report is available at ScholarlyCommons: https://repository.upenn.edu/cis_reports/479 Jack 5 User's Guide MS-CIS-91-78 GRAPHICS LAB 43 Cary B. Phillips Department of Computer and Information Science School of Engineering and Applied Science University of Pennsylvania Philadelphia, PA 19104-6389 October 1991 Jack @ User's Guide Version 5.8 January 11, 1994 Computer Graphics Research Laboratory Department of Computer and Information Sciences University of Pennsylvania Philadelphia, Pennsylvania Copyright @ 1988, 1992, 1993, 1994 University of Pennsylvania Jack @ is a registered trademark of the University of Pennsylvania. -
High-End Supercomputing at NASA!
High-End Supercomputing at NASA! Dr. Rupak Biswas! Chief, NASA Advanced Supercomputing (NAS) Division! Manager, High End Computing Capability (HECC) Project! NASA Ames Research Center, Moffett Field, Calif., USA ! 41st HPC User Forum Houston, TX 7 April 2011 NASA Overview: Mission Directorates •" NASA’s Mission: To pioneer the future in space exploration, scientific discovery, and aeronautics research •" Aeronautics Research (ARMD): Pioneer new flight technologies for safer, more secure, efficient, and environmentally friendly air transportation and space exploration •" Exploration Systems (ESMD): Develop new spacecraft and other capabilities for affordable, sustainable human and robotic exploration •" Science (SMD): Observe and understand the Earth-Sun system, our solar system, and the universe •" Space Operations (SOMD): Extend the duration and boundaries of human spaceflight for space exploration and discovery 2 NASA Overview: Centers & Facilities Goddard Inst for Space Studies Glenn Research Center New York, NY Cleveland, OH Headquarters Ames Research Center Washington, DC Moffett Field, CA Independent V&V Facility Goddard Space Flight Center Fairmont, WV Greenbelt, MD Dryden Flight Research Center Marshall Space Flight Center Langley Research Center Edwards, CA Huntsville, AL Hampton, VA White Sands Test Facility Wallops Flight Facility Jet Propulsion Laboratory White Sands, NM Stennis Space Center Wallops Island, VA Pasadena, CA Stennis, MS Johnson Space Center Kennedy Space Center Houston, TX Cape Canaveral, FL 3 M&S Imperative -
Scientific Application-Based Performance Comparison of SGI Altix 4700, IBM POWER5+, and SGI ICE 8200 Supercomputers
NAS Technical Report NAS-09-001, February 2009 Scientific Application-Based Performance Comparison of SGI Altix 4700, IBM POWER5+, and SGI ICE 8200 Supercomputers Subhash Saini, Dale Talcott, Dennis Jespersen, Jahed Djomehri, Haoqiang Jin, and Rupak Biswas NASA Advanced Supercomputing Division NASA Ames Research Center Moffett Field, California 94035-1000, USA {Subhash.Saini, Dale.R.Talcott, Dennis.Jespersen, Jahed.Djomehri, Haoqiang.Jin, Rupak.Biswas}@nasa.gov Abstract—The suitability of next-generation high-performance whereas the 4700 is based on the dual-core Itanium processor. computing systems for petascale simulations will depend on Hoisie et al. conducted performance comparison through various performance factors attributable to processor, memory, benchmarking and modeling of three supercomputers: IBM local and global network, and input/output characteristics. In this Blue Gene/L, Cray Red Storm, and IBM Purple [6]. Purple is paper, we evaluate performance of new dual-core SGI Altix 4700, an Advanced Simulation and Computing (ASC) system based quad-core SGI Altix ICE 8200, and dual-core IBM POWER5+ on the single-core IBM POWER5 architecture and is located at systems. To measure performance, we used micro-benchmarks Lawrence Livermore National Laboratory (LLNL) [7]. The from High Performance Computing Challenge (HPCC), NAS paper concentrated on system noise, interconnect congestion, Parallel Benchmarks (NPB), and four real-world applications— and performance modeling using two applications, namely three from computational fluid dynamics (CFD) and one from SAGE and Sweep3D. Oliker et al. studied scientific application climate modeling. We used the micro-benchmarks to develop a controlled understanding of individual system components, then performance on candidate petascale platforms: POWER5, analyzed and interpreted performance of the NPBs and AMD Opteron, IBM BlueGene/L, and Cray X1E [8].