Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Reducing Communication Costs for Sparse Matrix Multiplication within Algebraic Multigrid

Technical Report ·
DOI:https://doi.org/10.2172/1504845· OSTI ID:1504845
 [1];  [1];  [1]
  1. Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)

We consider the sequence of sparse matrix-matrix multiplications performed during the setup phase of algebraic multigrid. In particular, we show that the most commonly used parallel algorithm is often not the most communication-efficient one for all of the matrix-matrix multiplications involved. By using an alternative algorithm, we show that the communication costs are reduced (in theory and practice), and we demonstrate the performance benefit for both model (structured) and more realistic unstructured problems on large-scale distributed-memory parallel systems. Our theoretical analysis shows that we can reduce communication by a factor of up to 5.4 for a model problem, and we observe in our empirical evaluation communication reductions of factors up to 4.7 for structured problems and 3.7 for unstructured problems. These reductions in communication translate to run-time speedups of up to factors of 2.3 and 2.5, respectively.

Research Organization:
Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States); Sandia National Laboratories (SNL-CA), Livermore, CA (United States)
Sponsoring Organization:
USDOE National Nuclear Security Administration (NNSA)
DOE Contract Number:
AC04-94AL85000
OSTI ID:
1504845
Report Number(s):
SAND--2015-3275; 603395
Country of Publication:
United States
Language:
English

Similar Records

Communication-Avoiding Parallel Sparse-Dense Matrix-Matrix Multiplication
Conference · Sun May 01 00:00:00 EDT 2016 · Proceedings - 2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS) · OSTI ID:1769300

Reducing Parallel Communication in Algebraic Multigrid through Sparsification
Journal Article · Thu Dec 31 23:00:00 EST 2015 · SIAM Journal on Scientific Computing · OSTI ID:1808755

Distance-Two Interpolation for Parallel Algebraic Multigrid
Journal Article · Tue May 08 00:00:00 EDT 2007 · Numerical Linear Algebra with Applications, vol. 15, no. 2-3, December 3, 2007, pp. 115-139 · OSTI ID:940875

Related Subjects