Frontiers of Supercomputing II


 collapse sectionWelcome
 Supercomputing As a National Critical Technologies Effort
 collapse sectionGoals for Frontiers of Supercomputing II and Review of Events since 1983
 1983 Conference Summary
 Events in Supercomputing since 1983
 Conference Goals
 collapse sectionCurrent Status of Supercomputing in the United States
 The Global Imperative
 Importance of Computers—The Knowledge Economy
 Computers—A Historic Perspective
 collapse sectionCorrective Action
 Human Resources
 R&D Investment
 Technology Strategy
 International Cooperation

 collapse section2—  TECHNOLOGY PERSPECTIVE
 Supercomputing Tools and Technology
 High-Performance Optical Memory Technology at MCC
 Digital Superconductive Electronics
 collapse sectionEnabling Technology:  Photonics
 A Thousand Interconnections, Each at One Gigabit per Second
 One Connection at One Terabit per Second

 Vector Architecture in the 1990s
 In Defense of the Vector Computer
 Market Trends in Supercomputing
 Massively Parallel SIMD Computing on Vector Machines Using PASSWORK
 Vectors Are Different

 collapse section4—  SCALABLE PARALLEL SYSTEMS
 collapse sectionSymbolic Supercomputing
 Parallel Processing:  Moving into the Mainstream
 It's Time to Face Facts
 Large-Scale Systems and Their Limitations
 collapse sectionA Scalable, Shared-Memory, Parallel Computer
 Looking at All of the Options

 collapse section5—  SYSTEMS SOFTWARE
 Parallel Software
 collapse sectionSupercomputer Systems-Software Challenges
 Distributed Computing
 High-Speed Networks
 Virtual Memory
 Resource Management
 Parallel Processing
 Future Supercomputing Elements
 Compiler Issues for TFLOPS Computing
 collapse sectionPerformance Studies and Problem-Solving Environments
 Systems and Software

 collapse section6—  USER-INTERFACE SOFTWARE
 Parallel Architecture and the User Interface
 collapse sectionObject-Oriented Programming, Visualization, and User-Interface Issues
 collapse sectionObject-Oriented Parallel Programming
 Distributed Computing
 Data Parallel Programming
 Visualization Requirements
 collapse sectionSoftware Issues at the User Interface
 collapse sectionCompilers and Communication
 Myrias SPS-2:  Virtual Memory on a Distributed System
 Myrias SPS-2:  A Concrete Example
 Myrias SPS-2:  Efficiency of Virtual Memory
 The Connection Machine CM-2:  Overlapping Communication with Computation
 Debugging Tools
 High-Level Languages, Extensions, Libraries, and Graphics
 Future Supercomputing Environments:  Heterogeneous Systems
 An Application for a Heterogeneous System
 What Can We Learn from Our Experience with Parallel Computation up to Now?

 collapse sectionParallel Algorithms and Implementation Strategies on Massively Parallel Supercomputers
 Some Developments in Parallel Algorithms
 Some Developments in Parallel Applications
 Some Developments in Parallel Applications II
 Closing Remarks
 The Interplay between Algorithms and Architectures:  Two Examples
 collapse sectionLinear Algebra Library for High-Performance Computers
 collapse sectionLINPACK Benchmark
 Transfer Rate
 Memory Latency
 Development of Standards
 collapse sectionAlgorithm Design
 Divide-and-Conquer Approach
 Future Directions for Research
 Design of Algorithms
 Computing for Correctness

 Interactive Steering of Supercomputer Calculations
 A Vision of the Future at Sun Microsystems
 On the Future of the Centralized Computing Environment
 collapse sectionMolecular Nanotechnology
 collapse sectionSupercomputing Alternatives
 Less Is More
 Supersubstitutes Provide More Overall Capacity
 collapse sectionHow Supers Are Being Niched
 Minicomputers and Superminis
 Massively Data-Parallel Computers
 Why Supercomputers Are Becoming Less General Purpose
 The Supercomputer Industry
 Is the Supercomputer Industry Hastening Its Own Demise?
 A Smaller, Healthier Supercomputer Industry
 collapse sectionPolicy Issues
 Supporting Circuit and Packaging Technology
 Supers and Security
 Supers for Competitiveness
 Epilogue, June 1992

 collapse sectionOverview of Industrial Supercomputing
 Why Use Supercomputing at All?
 Impediments to Industrial Use of Supercomputers
 Technology Transfer and Collaboration
 Shell Oil Supercomputing
 Government's High Performance Computing Initiative Interface with Industry
 collapse sectionAn Overview of Supercomputing at General Motors Corporation
 People and the Machine Environment
 History of Supercomputing at GM
 Automotive Industry Interest in Supercomputers
 Long-Term Benefits
 Needs and Challenges
 Barriers to Use of Supercomputers in the Industrial Environment

 Planning for a Supercomputing Future
 collapse sectionHigh-Performance Computing at the National Security Agency
 Characterization of HPC
 HPC Architecture
 Software Environment
 Mass-Storage Requirements
 Summary of Issues
 collapse sectionThe High Performance Computing Initiative:  A Way to Meet NASA's Supercomputing Requirements for Aerospace
 The Role of Computing in National Defense Technology
 NSF Supercomputing Program

 collapse section11—  INTERNATIONAL ACTIVITY
 collapse sectionA Look at Worldwide High-Performance Computing and Its Economic Implications for the U.S.
 collapse sectionA Brief Technical Overview of the Present-Day Landscape
 The Soviet Union
 Western Europe
 collapse sectionThe Japanese Challenge and "McAdams's Laws"
 collapse sectionIntroduction
 Law 1—  That Which Is Currently Taking Place Is Not Impossible
 Japan:  Vertical Integration, Keiretsu, and Government Coordination
 collapse sectionThe U.S.:  Rugged Individualism and Trade-War Losses
 Law 2—  You Don't Catch up without Catching Up
 collapse sectionTrade:  "Successful" Negotiations and "Potato Chips"
 Law 3—  When Two Countries Are in a Trade War and One Does Not Realize It, That Country Is Unlikely to Win
 collapse sectionRemedies
 Law 4—  An Important Aspect of Change Is That Things Are Different Afterward
 The Future
 References and Bibliography
 collapse sectionEconomics, Revelation, Reality, and Computers

 Supercomputing since 1983
 collapse sectionLessons Learned
 Parallel Processing:  1980 to 2000
 The Attack of the Killer Micros
 Programmer Productivity on Massively Parallel Systems
 Front End/Back End Versus Native UNIX
 Single User versus Multiple Users
 Interconnect Performance, System Versatility, and Delivered Performance
 Challenges and Directions for the Future
 Appendix:  BBN Parallel-Processing Systems
 collapse sectionThe John von Neumann Computer Center:  An Analysis
 collapse sectionIntroduction
 The "Pre-Lax Report" Period
 The Lax Report
 Establishment of the Centers
 collapse sectionThe John von Neumann Center
 The Proposal
 Consortium for Scientific Computing
 The Universities
 The State of New Jersey
 The NSF
 Zero One
 collapse sectionWhat Went Wrong?
 The Analysis
 Corporate Problems
 NSF, Funding, and Funding Leverage
 Project THOTH:  An NSA Adventure in Supercomputing, 1984–88
 collapse sectionThe Demise of ETA Systems
 In the Beginning
 Industry Observations
 FPS Computing:  A History of Firsts

 collapse sectionWhy Supercomputing Matters:  An Analysis of the Economic Impact of the Proposed Federal High Performance Computing Initiative
 collapse sectionPhase I Methodology
 Scenario A
 Scenario B
 Phase II Methodology
 Government As Buyer and Leader
 Concerns about Policies and Economics for High-Performance Computing
 High-Performance Computing in the 1990s
 A High-Performance Computing Association to Help the Expanding Supercomputing Industry
 collapse sectionThe New Supercomputer Industry
 The View from DEC
 Industry Perspective:  Remarks on Policy and Economics for High-Performance Computing

 collapse section14—  WHAT NOW?
 collapse sectionConference Summary
 Skilled, Imaginative Users and a Broad Spectrum of Applications
 Workstations and Visualization Engines
 Mass Storage and Accessible Knowledge Bases
 Heterogeneous High-Performance Computer Engines
 Fast, Local, Wide-Area, and National Networks
 Software Environment
 Concluding Remarks
 The High Performance Computing Initiative
 Government Bodies As Investors
 Realizing the Goals of the HPCC Initiative:  Changes Needed
 The Importance of the Federal Government's Role in High-Performance Computing
 Legislative and Congressional Actions on High-Performance Computing and Communications
 The Federal Role As Early Customer
 A View from the Quarter-Deck at the National Security Agency
 Supercomputers and Three-Year-Olds
 NASA's Use of High-Performance Computers:  Past, Present, and Future
 A Leadership Role for the Department of Commerce


collapse section Collapse All | Expand All expand section