Skip to Main content Skip to Navigation
Conference papers

Analysis of Partitioning Models and Metrics in Parallel Sparse Matrix-Vector Multiplication

Abstract : Graph/hypergraph partitioning models and methods have been successfully used to minimize the communication among processors in several parallel computing applications. Parallel sparse matrix-vector multiplication (SpMxV) is one of the representative applications that renders these models and methods indispensable in many scientific com- puting contexts. We investigate the interplay of the partitioning metrics and execution times of SpMxV implementations in three libraries: Trilinos, PETSc, and an in-house one. We carry out experiments with up to 512 processors and investigate the results with regression analysis. Our experiments show that the partitioning metrics influence the perfor- mance greatly in a distributed memory setting. The regression analyses demonstrate which metric is the most influential for the execution time of the libraries.
Complete list of metadatas

Cited literature [15 references]  Display  Hide  Download

https://hal.inria.fr/hal-00923454
Contributor : Equipe Roma <>
Submitted on : Thursday, January 2, 2014 - 7:58:34 PM
Last modification on : Wednesday, November 20, 2019 - 3:18:03 AM
Long-term archiving on: : Saturday, April 8, 2017 - 10:34:22 AM

File

spmxvLNCS.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Kamer Kaya, Bora Uçar, Umit Catalyurek. Analysis of Partitioning Models and Metrics in Parallel Sparse Matrix-Vector Multiplication. 10th PPAM - Parallel Processing and Applied Mathematics, Sep 2013, Varsovie, Poland. pp.174--184, ⟨10.1007/978-3-642-55195-6_16⟩. ⟨hal-00923454⟩

Share

Metrics

Record views

884

Files downloads

773