S10: Scaling to Petascale and Beyond: Performance Analysis and Optimization of Applications
Sunday, Nov. 13
8:30-5:00
Glenn Brook, University of Tennessee, Knoxville
Donald Frederick, Lawrence Livermore National Laboratory
Richard Gerber, NERSC / Lawrence Berkeley National Laboratory
Jeff Larkin, Cray, Inc.
http://sc11.supercomputing.org/schedule/event_detail.php?evid=tut146
ABSTRACT:
Current supercomputer systems have architectures with many thousands of cores. Application developers are faced with the task of designing for parallel scalability at ever higher levels. The effective use of such platforms, requires application developers to focus on achieving parallel scalability through application design and optimization driven by performance analysis. This tutorial provides developers with the background in system features, program design and performance analysis to aid in producing efficient programs on modern supercomputers.
AGENDA:
Design Considerations
8:30 - Introduction to Petascale Systems – Richard Gerber
8:50 - HPC Communication - Glenn Brook
9:45 - HPC IO & Lustre File Systems - Richard Gerber
10:00 - Break
10:20 - HPC IO & Lustre File Systems Continued - Richard Gerber
11:15 - HPC Computation - 40 minutes - Jeff Larkin
Lunch (12:00-1:15)
Multi-Core Performance Analysis
1:15 - Introduction and General Methodology - Don Frederick
1:35 - Profiling with Open SpeedShop - Don Frederick
Extreme Scaling on Petascale-class Systems
2:20 - Cray XT Architecture, System Software, and Scientific Libraries - Jeff Larkin
2:35 - Cray XT Porting, Scaling, and Optimization - Jeff Larkin
3:00 - Break
3:15 - Cray XT Porting, Scaling, and Optimization Continued
4:00 - IBM Blue Gene P Architecture, System Software, and Scientific Libraries - Don Frederick
4:15 - IBM Blue Gene P Porting, Scaling and Optimization Case Studies – Don Frederick
5:00 - Adjourn
Downloads
-
IntrotoPetascaleDF-Perf.11.11.2.pdf | Adobe Acrobat PDF fileIntroduction to Performance Analysis for HPC, Don Frederick, LLNL
-
IntrotoPetascaleDF-OSS.11.12.11.pdf | Adobe Acrobat PDF fileHow to use Open|SpeedShop to Analyze the Performance of Parallel Codes, Don Frederick, LLNL
-
larkincrayswhw.pdf | Adobe Acrobat PDF fileCray XE/XT Architecture, System Software, and Scientific Libraries, Jeff Larkin, Cray, Inc.
-
larkinoptimization2.pdf | Adobe Acrobat PDF fileCray XE/XT Porting, Scaling, and Optimization, Jeff Larkin, Cray
-
IntrotoPetascaleDF-BG.11.11.11.pdf | Adobe Acrobat PDF fileIBM Blue Gene Architecture, Don Frederick, LLNL
-
IntrotoPetascaleDF-ddcMD.11.14.11.pdf | Adobe Acrobat PDF fileCase Case Study DDCMP: Beyond Homogeneous Decomposition with ddcMD, Don Frederick, LLNL
-
IntrotoPetascaleDF-QBOX.11.14.11.pdf | Adobe Acrobat PDF fileCase Study: Beyond Homogeneous Decomposition with Qbox, Don Frederick, LLNL