Abstract | ||
---|---|---|
Conjugate gradient (CG) solvers are the most time consuming part of many scientific applications. These solvers exhibit communication operations that can prevent high performance from being achieved on large-scale systems. In this paper we present a novel technique to boost the performance of these solvers. In this, multiple independent solvers, which occur in some applications, are combined allowing for the overlapping of communication with other communication and computation, resulting with increased performance. This is the first work where a combination of CG solvers has been exploited and offers performance improvements, which may be particularly important in very large-scale systems. Results are presented for the MIMD lattice computation (MILC) application that show the cost of collective communications can be reduced by a factor of up to 2.5×. Moreover the performance of MILC is significantly improved, by over 10% for typical lattice sizes on a 1024-processor system, and by 15% on a 4096-processor system. Larger improvements are expected on larger systems. Copyright © 2009 John Wiley & Sons, Ltd. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1002/cpe.v21:14 | Concurrency and Computation: Practice and Experience |
Keywords | Field | DocType |
performance improvement,1024-processor system,collective communication,multiple independent solvers,high performance,solvers exhibit communication operation,cg solvers,multiple conjugate gradient solvers,4096-processor system,increased performance,large-scale system,iteration method,conjugate gradient,parallel processing | Conjugate gradient method,Lattice (order),Iterative method,Computer science,Parallel computing,Parallel processing,Computation,MIMD | Journal |
Volume | Issue | ISSN |
21 | 14 | 1532-0626 |
Citations | PageRank | References |
0 | 0.34 | 3 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
José Carlos Sancho | 1 | 382 | 29.97 |
Darren J. Kerbyson | 2 | 1102 | 104.36 |