Title :
A nested dissection partitioning method for parallel sparse matrix-vector multiplication
Author :
Boman, Erik G. ; Wolf, Michael M.
Author_Institution :
Sandia Nat. Labs., Albuquerque, NM, USA
Abstract :
We consider how to map sparse matrices across processes to reduce communication costs in parallel sparse matrix-vector multiplication, an ubiquitous kernel in high performance computing. Our main contributions are: (i) an exact graph model for communication with general (two-dimensional) matrix distribution, and (ii) a recursive partitioning algorithm based on nested dissection that approximately solves this model. We have implemented our algorithm using hypergraph partitioning software to enable a fair comparison with existing methods. We present partitioning results for sparse structurally symmetric matrices from several application areas. Our new method is competitive with the best 2D algorithm (fine-grain hypergraph model) in terms of communication volume, but requires fewer messages. The nested dissection method is almost as fast to compute as 1D methods and the communication volume is significantly reduced (up to 97%) compared to 1D layout. Further improvements in quality may be possible by small modifications to existing nested dissection ordering software.
Keywords :
graph theory; mathematics computing; matrix multiplication; parallel algorithms; sparse matrices; ubiquitous computing; vectors; 2D algorithm; communication volume; exact graph model; high performance computing; hypergraph model; hypergraph partitioning software; nested dissection partitioning method; parallel sparse matrix-vector multiplication; recursive partitioning algorithm; sparse matrix mapping; sparse structurally symmetric matrices; two-dimensional matrix distribution; ubiquitous kernel; Particle separators; Partitioning algorithms; Software algorithms; Solid modeling; Sparse matrices; Symmetric matrices; Vectors;
Conference_Titel :
High Performance Extreme Computing Conference (HPEC), 2013 IEEE
Conference_Location :
Waltham, MA
Print_ISBN :
978-1-4799-1364-0
DOI :
10.1109/HPEC.2013.6670333