Name
Title | ||
|---|---|---|
Exploiting hierarchy parallelism for molecular dynamics on a petascale heterogeneous system |
Abstract | ||
|---|---|---|
Heterogeneous systems with nodes containing more than one type of computation units, e.g., central processing units (CPUs) and graphics processing units (GPUs), are becoming popular because of their low cost and high performance. In this paper, we have developed a Three-Level Parallelization Scheme (TLPS) for molecular dynamics (MD) simulation on heterogeneous systems. The scheme exploits multi-level parallelism combining (1) inter-node parallelism using spatial decomposition via message passing, (2) intra-node parallelism using spatial decomposition via dynamically scheduled multi-threading, and (3) intra-chip parallelism using multi-threading and short vector extension in CPUs, and employing multiple CUDA threads in GPUs. By using a hierarchy of parallelism with optimizations such as communication hiding intra-node, and memory optimizations in both CPUs and GPUs, we have implemented and evaluated a MD simulation on a petascale heterogeneous supercomputer TH-1A. The results show that MD simulations can be efficiently parallelized with our TLPS scheme and can benefit from the optimizations. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1016/j.jpdc.2013.07.015 | J. Parallel Distrib. Comput. |
Keywords | Field | DocType |
intra-node parallelism,inter-node parallelism,intra-chip parallelism,memory optimizations,petascale heterogeneous system,molecular dynamic,heterogeneous system,multi-level parallelism,md simulation,petascale heterogeneous supercomputer th-1a,tlps scheme,spatial decomposition,exploiting hierarchy parallelism,molecular dynamics,gpu computing | Instruction-level parallelism,Supercomputer,CUDA,Task parallelism,Computer science,Parallel computing,Data parallelism,General-purpose computing on graphics processing units,Petascale computing,Message passing | Journal |
Volume | Issue | ISSN |
73 | 12 | 0743-7315 |
Citations | PageRank | References |
6 | 0.47 | 25 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Qiang Wu | 1 | 68 | 6.85 |
| Canqun Yang | 2 | 188 | 29.39 |
| Tao Tang | 3 | 42 | 7.44 |
| Liquan Xiao | 4 | 106 | 15.43 |