Paper Info
Title
Performance study of matrix computations using multi-core programming tools
Abstract
Basic matrix computations such as vector and matrix addition, dot product, outer product, matrix transpose, matrix - vector and matrix multiplication are very challenging computational kernels arising in scientific computing. In this paper, we parallelize those basic matrix computations using the multi-core and parallel programming tools. Specifically, these tools are Pthreads, OpenMP, Intel Cilk++, Intel TBB, Intel ArBB, SMPSs, SWARM and FastFlow. The purpose of this paper is to present an unified quantitative and qualitative study of these tools for parallel matrix computations on multicore. Finally, based on the performance results with compilation optimization we conclude that the Intel ArBB and SWARM parallel programming tools are the most appropriate because these give good performance and simplicity of programming. In particular, we conclude that the Intel ArBB is a good choice for implementing intensive computations such as matrix product because it gives significant speedup results over the serial implementation. On the other hand, the SWARM tool gives good performance results for implementing matrix operations of medium size such as vector addition, matrix addition, outer product and matrix - vector product.
Year
DOI
Venue
2012
10.1145/2371316.2371353
BCI
Keywords
Field
DocType
intel cilk,intel arbb,matrix transpose,parallel matrix computation,matrix operation,multi-core programming tool,matrix multiplication,basic matrix,performance study,outer product,matrix product,matrix addition,multi core,parallel programming,linear algebra
Matrix addition,Transpose,Matrix (mathematics),Computer science,Parallel computing,Computational science,Dot product,Cilk,Multi-core processor,Matrix multiplication,Speedup
Conference
Citations 
PageRank 
References 
0
0.34
10
Authors
2
Name
Order
Citations
PageRank
Panagiotis D. Michailidis16011.16
Konstantinos G. Margaritis230345.46