Performance Analysis of Blue Gene/L Using Parallel Discrete Event Simulation
Abstract
High performance computers currently under construction, such as IBM's Blue Gene/L, consisting of large numbers (64K) of low cost processing elements with relatively small local memories (256MB) connected via relatively low bandwidth (0.0625 Bytes/FLOP) low cost interconnection networks promise exceptional cost-performance for some scientific applications. Due to the large number of processing elements and adaptive routing networks in such systems, performance analysis of meaningful application kernels requires innovative methods. This paper describes a method that combines application analysis, tracing and parallel discrete event simulation to provide early performance prediction. Specifically, results of performance analysis of a Lennard-Jones Spatial (LJS) Decomposition molecular dynamics benchmark code for Blue Gene/L are given.
Attached Files
Submitted - BG_L_Technical_Paper_CACR-2003-194.pdf
Files
Name | Size | Download all |
---|---|---|
md5:01cb97fbf26bfa06bafdecc0fadb1071
|
721.8 kB | Preview Download |
Additional details
- Eprint ID
- 28172
- Resolver ID
- CaltechCACR:CACR-2003-194
- Created
-
2004-04-01Created from EPrint's datestamp field
- Updated
-
2019-10-09Created from EPrint's last_modified field
- Caltech groups
- Center for Advanced Computing Research