This interesting book of conference proceedings can be read online here: https://publishing.cdlib.org/ucpressebooks/view?docId=ft0f59n73z;brand=ucpress
Set in ’93 being 10 years after the first Frontiers conference the talks capture the key topics during the early stages of the massively parallel processing era. Articles by Les Davis and Bo Ewald and others.
Frontiers of Supercomputing II
A National Reassessment
Edited by Karyn R. Ames and Alan Brenner
UNIVERSITY OF CALIFORNIA PRESS
Berkeley · Los Angeles · Oxford
© 1994 The Regents of the University of California
Introduction starts:
“Welcome to Los Alamos and to New Mexico. I think most of you know that it was in 1983—in fact, seven years ago this week—that we held the first Frontiers of Supercomputing conference here at Los Alamos under the sponsorship of Los Alamos National Laboratory and the National Security Agency (NSA) to assess the critical issues that face supercomputing. Today we are here to make a national reassessment of supercomputing.”
Preferred Citation: Ames, Karyn R., and Alan Brenner, editors Frontiers of Supercomputing II: A National Reassessment. Berkeley: University of California Press, c1994 1994. http://ark.cdlib.org/ark:/13030/ft0f59n73z/
Index
Frontiers of Supercomputing II
PREFACE
ACKNOWLEDGMENTS
1— OPENING, BACKGROUND, AND QUESTIONS POSED FOR THIS CONFERENCE
Welcome
* Reference
* Supercomputing As a National Critical Technologies Effort
Goals for Frontiers of Supercomputing II and Review of Events since 1983
* 1983 Conference Summary
* Events in Supercomputing since 1983
* Conference Goals
* Questions
Current Status of Supercomputing in the United States
* The Global Imperative
* Importance of Computers—The Knowledge Economy
* Computers—A Historic Perspective
Corrective Action
* Human Resources
* R&D Investment
* Technology Strategy
* International Cooperation
* Summary
2— TECHNOLOGY PERSPECTIVE
* Overview
* Supercomputing Tools and Technology
* High-Performance Optical Memory Technology at MCC
* Digital Superconductive Electronics
Enabling Technology: Photonics
* Introduction
* A Thousand Interconnections, Each at One Gigabit per Second
* One Connection at One Terabit per Second
* References
3— VECTOR PIPELINE ARCHITECTURE
* 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
4— SCALABLE PARALLEL SYSTEMS
Symbolic Supercomputing
* References
* Parallel Processing: Moving into the Mainstream
* It’s Time to Face Facts
* Large-Scale Systems and Their Limitations
A Scalable, Shared-Memory, Parallel Computer
* References
* Looking at All of the Options
5— SYSTEMS SOFTWARE
* Parallel Software
Supercomputer Systems-Software Challenges
* Abstract
* Introduction
* Distributed Computing
* High-Speed Networks
* Virtual Memory
* Resource Management
* Parallel Processing
* Progress
* Future Supercomputing Elements
* Compiler Issues for TFLOPS Computing
Performance Studies and Problem-Solving Environments
* References
* Systems and Software
6— USER-INTERFACE SOFTWARE
* Parallel Architecture and the User Interface
Object-Oriented Programming, Visualization, and User-Interface Issues
Object-Oriented Parallel Programming
* Distributed Computing
* Data Parallel Programming
* Visualization Requirements
* References
Software Issues at the User Interface
* Abstract
* Introduction
Compilers 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
* Conclusions
* References
* What Can We Learn from Our Experience with Parallel Computation up to Now?
7— ALGORITHMS FOR HIGH-PERFORMANCE COMPUTING
Parallel Algorithms and Implementation Strategies on Massively Parallel Supercomputers
* Introduction
* Some Developments in Parallel Algorithms
* Some Developments in Parallel Applications
* Some Developments in Parallel Applications II
* Closing Remarks
* References
* The Interplay between Algorithms and Architectures: Two Examples
Linear Algebra Library for High-Performance Computers
* Introduction
* LINPACK
LINPACK Benchmark
* Transfer Rate
* Memory Latency
* Development of Standards
* LAPACK
Algorithm Design
* Divide-and-Conquer Approach
* Accuracy
* Tools
* Testing
* Future Directions for Research
* References
* Design of Algorithms
* Computing for Correctness
8— THE FUTURE COMPUTING ENVIRONMENT
* Interactive Steering of Supercomputer Calculations
* A Vision of the Future at Sun Microsystems
* On the Future of the Centralized Computing Environment
Molecular Nanotechnology
* References
Supercomputing Alternatives
* Less Is More
* Supersubstitutes Provide More Overall Capacity
How Supers Are Being Niched
* Workstations
* Minicomputers and Superminis
* Mainframes
* Massively Data-Parallel Computers
* Minisupercomputers
* Superworkstations
* Why Supercomputers Are Becoming Less General Purpose
* The Supercomputer Industry
* Is the Supercomputer Industry Hastening Its Own Demise?
* A Smaller, Healthier Supercomputer Industry
Policy Issues
* Supporting Circuit and Packaging Technology
* Supers and Security
* Supers for Competitiveness
* Conclusions
* Epilogue, June 1992
9— INDUSTRIAL SUPERCOMPUTING
Overview of Industrial Supercomputing
* Abstract
* Introduction
* Why Use Supercomputing at All?
* Impediments to Industrial Use of Supercomputers
* Technology Transfer and Collaboration
* Conclusion
* References
* Shell Oil Supercomputing
* Government’s High Performance Computing Initiative Interface with Industry
An Overview of Supercomputing at General Motors Corporation
* Abstract
* Introduction
* People and the Machine Environment
* History of Supercomputing at GM
* Automotive Industry Interest in Supercomputers
* Applications
* Long-Term Benefits
* Needs and Challenges
* References
* Barriers to Use of Supercomputers in the Industrial Environment
10— GOVERNMENT SUPERCOMPUTING
* Planning for a Supercomputing Future
High-Performance Computing at the National Security Agency
* Introduction
* Characterization of HPC
* HPC Architecture
* Software Environment
* Mass-Storage Requirements
* Summary of Issues
The High Performance Computing Initiative: A Way to Meet NASA’s Supercomputing Requirements for Aerospace
* Reference
* The Role of Computing in National Defense Technology
* NSF Supercomputing Program
11— INTERNATIONAL ACTIVITY
A Look at Worldwide High-Performance Computing and Its Economic Implications for the U.S.
* Abstract
A Brief Technical Overview of the Present-Day Landscape
* The Soviet Union
* Western Europe
* Japan
The Japanese Challenge and “McAdams’s Laws”
Introduction
* Law 1— That Which Is Currently Taking Place Is Not Impossible
* Japan: Vertical Integration, Keiretsu, and Government Coordination
The U.S.: Rugged Individualism and Trade-War Losses
* Law 2— You Don’t Catch up without Catching Up
Trade: “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
Remedies
* Law 4— An Important Aspect of Change Is That Things Are Different Afterward
* The Future
* References and Bibliography
Economics, Revelation, Reality, and Computers
* References
12— EXPERIENCE AND LESSONS LEARNED
* Supercomputing since 1983
Lessons Learned
* Abstract
* Introduction
* 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
* Summary
* Appendix: BBN Parallel-Processing Systems
* References
The John von Neumann Computer Center: An Analysis
Introduction
* The “Pre-Lax Report” Period
* The Lax Report
* Establishment of the Centers
The John von Neumann Center
* The Proposal
* Consortium for Scientific Computing
* The Universities
* The State of New Jersey
* The NSF
* ETA
* Zero One
* JVNC
What Went Wrong?
* The Analysis
* Location
* Corporate Problems
* NSF, Funding, and Funding Leverage
* Governance
* Conclusions
* Project THOTH: An NSA Adventure in Supercomputing, 1984–88
The Demise of ETA Systems
* In the Beginning
* Hardware
* Software
* Industry Observations
* FPS Computing: A History of Firsts
13— INDUSTRY PERSPECTIVE: POLICY AND ECONOMICS FOR HIGH-PERFORMANCE COMPUTING
Why Supercomputing Matters: An Analysis of the Economic Impact of the Proposed Federal High Performance Computing Initiative
* Introduction
Phase 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
The New Supercomputer Industry
* Reference
* The View from DEC
* Industry Perspective: Remarks on Policy and Economics for High-Performance Computing
14— WHAT NOW?
Conference Summary
* Introduction
* 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
* Farewell