Abstract | ||
---|---|---|
The recent advent of parallel machines with tens of thousands of processors is presenting new challenges for obtaining scalability. A particular challenge for large-scale scientific software is determining the inter-processor communications required by the computation when a global description of the data is unavailable or too costly to store. We present a type of rendezvous algorithm that determines communication partners in a scalable manner by assuming the global distribution of the data. We analyze the algorithm theoretically and demonstrate its scaling properties on up to 32,768 processors of BlueGene/L in the context of determining communication patterns for a matrix-vector multiply in the hypre software library. Our algorithm is very general and is applicable to a variety of situations in parallel computing. |
Year | DOI | Venue |
---|---|---|
2006 | 10.1016/j.parco.2006.06.009 | Parallel Computing |
Keywords | Field | DocType |
global partition,global distribution,rendezvous algorithm,scalability,algorithm theoretically,processor inter-communication,communication pattern,global description,parallel computing,distributed directory,hypre software library,large-scale scientific software,communication partner,assumed partition algorithm,inter-processor communication,parallel computer | Partition problem,Scientific software,Computer science,Parallel computing,Theoretical computer science,Software,Rendezvous,Scaling,Computation,Distributed computing,Scalability | Journal |
Volume | Issue | ISSN |
32 | 5 | Parallel Computing |
Citations | PageRank | References |
8 | 0.81 | 8 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Allison H. Baker | 1 | 222 | 15.49 |
R. D. Falgout | 2 | 102 | 16.21 |
Ulrike Meier | 3 | 51 | 19.82 |