The Alliance Distributed Supercomputing Facilities

1/1/99


Click here to start


Table of Contents

The Alliance Distributed Supercomputing Facilities

The National PACI Program - Partners and Supercomputer Users

PACI - The NSF Partnerships for Advanced Computational Infrastructure

NCSA is Combining Shared Memory Programming with Massive Parallelism

NCSA Users by System

Millions of NUs Used at NCSA FY93 to FY98

NCSA Supplies Cycles to a Logarithmic Demand Function of Projects

Evolution of the NCSA Project Distribution Function FY93 to FY98

Rapid Increase in Large projects at NCSA FY93-98

Breakout in Supporting Super Projects at NCSA in the Last Year

Migration of NCSA User Distribution Toward the High End

Alliance LES Chose 27 Large PSC Projects to Track Out of 100 Targeted Projects

Disciplines Represented in the Large Academic Projects at the Alliance LES

Application Performance Scaling on 128-Processor Origin

Origin Brings Shared Memory to MPP Scalability

The Growth Rate of the National Capacity is Slowing Down Again

The Drop in High End Capacity Available to National Academic Researchers

Major Gap Has Developed in National Usage at NSF Supercomputer Centers

Allocated Capacity for Meritorious NSF Large National Projects

Clustered Shared Memory Computers are Today’s High End

High-End Computing Enables High Resolution of Flow Details

Cycles Used by NSF Community at the NSF Supercomputer Centers by Vendor

Peak Teraflops in Aggressive Upgrade Plan

Deputy Director Bordogna on NSF Leadership in Information Technologies

President’s Information Technology Advisory Committee Interim Report

PITAC Draft Refinement of High-End Acquisition Recommendation

Harnessing the Unused Cycles of Networks of Workstations

NT Workstation Shipments Rapidly Surpassing UNIX

PACI Fostering Commodity Computing

Performance Analysis is Key Computer Science Research Enabling Computational Science

Performance of Scalable Systems Shows the Promise of Local Clustered PCs

Near Perfect Scaling of Cactus - 3D Dynamic Solver for the Einstein GR Equations

QCD Performance on Various Machines

The Road to Intel’s Merced The Convergence of Scientific and Commercial Computing

The NCSA Information Workbench - An Architecture for Web-Based Computing

Using a Web Browser to Run Programs and Analyze Data Worldwide

Author: Larry Smarr

Email: lsmarr@ncsa.uiuc.edu

Home Page: http://www.ncsa.uiuc.edu/People/pls

Download presentation source