Table of Contents
Parallel Debugging and Performance Analysis Tools: NetSolve,Vampir, and TotalView
High-Performance Computing, Trends, and Grid Based Computing
Outline
Computer Science Department Computer Science & Mathematics Division
Technology Trends: Microprocessor Capacity
PPT Slide
PPT Slide
PPT Slide
TOP500 Technical Systems by Vendor
PPT Slide
The Maturation of Highly Parallel Technology
Hardware Framework: Clusters
Capability Software (1)
Capability Software (2)
High-Performance Computing Directions
Future: Petaflops ( fl pt ops/s)
Two Designs for a Petaflops System
Two Designs for a Petaflops System (Continued)
Petaflops: Key Algorithm Research Issues
How To Get Performance From Commodity Processors?
Why ATLAS is needed
Adaptive Approach for Level 3 BLAS
Code Generation Strategy
ATLAS DGEMM across platforms, 500x500 problem
500 x 500 Double Precision Right-Looking LU factorization
Multithreaded BLAS for Performance
Algorithmic approach for Level 3 BLAS
ATLAS and the Computational Grid
Motivation for Grid Computing
Problem Solving Environments & Computational Grid
Vision for the Grid
PPT Slide
Network Enabled Servers
Motivation for NetSolve
PPT Slide
PPT Slide
PPT Slide
NetSolve - MATLAB Interface
PPT Slide
PPT Slide
PPT Slide
PPT Slide
NetSolve Applications and Interactions
Hiding the Parallel Processing
MCell: 3-D Monte-Carlo Simulation of Neuro-Transmitter Release in Between Cells
NetSolve - Integrating ImageVision, Remote Image Processing and Computational Steering
PPT Slide
Technology Integration
NetSolve and Globus
NetSolve-Globus: Design
NetSolve-Globus: Status
PPT Slide
PPT Slide
PPT Slide
PPT Slide
PPT Slide
PPT Slide
NetSolve: Features for the Future
Next Step: Hardware & Software Servers
NetSolve Available Today
Conclusion
Contributors to These Ideas
|