Parallel Sparse Matrix-Matrix Multiplication and Indexing: Implementation and Experiments

Details Parallel Sparse Matrix-Matrix Multiplication and Indexing: Implementation and Experiments Parallel Sparse Matrix-Matrix Multiplication and Indexing: Implementation and Experiments

2011-09-16

arxiv.org/abs/1109.3739v1

Computer Science

Distributed, Parallel, and Cluster Computing

Scientific paper

Generalized sparse matrix-matrix multiplication (or SpGEMM) is a key primitive for many high performance graph algorithms as well as for some linear solvers, such as algebraic multigrid. Here we show that SpGEMM also yields efficient algorithms for general sparse-matrix indexing in distributed memory, provided that the underlying SpGEMM implementation is sufficiently flexible and scalable. We demonstrate that our parallel SpGEMM methods, which use two-dimensional block data distributions with serial hypersparse kernels, are indeed highly flexible, scalable, and memory-efficient in the general case. This algorithm is the first to yield increasing speedup to an unbounded number of processors; our experiments show scaling up to thousands of processors in a variety of test scenarios.

Affiliated with

Also associated with

No associations

Rating

Parallel Sparse Matrix-Matrix Multiplication and Indexing: Implementation and Experiments does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Parallel Sparse Matrix-Matrix Multiplication and Indexing: Implementation and Experiments, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Parallel Sparse Matrix-Matrix Multiplication and Indexing: Implementation and Experiments will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-142196