Task-Queue Based Hybrid Parallelism: A Case Study
Additional information
Authors
Fürlinger K.,
Schenk O.,
Hagemann M.
Type
Article in conference proceedings
Year
2004
Language
English
Abstract
In this paper we report on our experiences with hybrid parallelism in PARDISO, a high-performance sparse linear solver. We start with the OpenMP-parallel numerical factorization algorithm and re-organize it using a central dynamic task queue to be able to add message passing functionality. The hybrid version allows the solver to run on a larger number of processors in a cost effective way with very reasonable performance. A speed-up of more than nine running on a four-node quad Itanium 2 SMP cluster is achieved in spite of the fact that a large potential to minimize MPI communication is not yet exploited in the first version of the implementation.
Conference proceedings
proceedings of the European Conference on Parallel Processing
Publisher
Springer Berlin Heidelberg
Start page number
624
End page number
631
Meeting name
Euro-Par 2004: Euro-Par 2004 Parallel Processing
Meeting place
Pisa, Italy
Meeting date
August 31, 2004 - September 03, 2004
Editor
Springer, Berlin, Heidelberg