===firstname:
Amanda
===firstname3:
Jon
===affil6:

===lastname3:
Calhoun
===email:
bienz2@illinois.edu
===keyword_other2:

===lastname6:

===affil5:

===lastname4:
Gropp
===lastname7:

===affil7:

===postal:
202 N Race St
Apt 312
Urbana, IL 61801
===ABSTRACT:
In many sparse iterative methods, the sparse matrix-vector multiply (SpMV) is the dominant computational kernel.  This is particularly true in a parallel setting where communication costs increase dramatically at large core counts.  Both the structure and number of nonzeros impact the communication patterns and we introduce several strategies in this talk to limit the impact on performance.

While many current methods target a reduction in communication requirements, we can additionally hide the cost of communication through asynchrony.  Our approach decomposes the monolithic communication in standard approaches of the SpMV by pipelining segments of the data.  We also explore over-decomposition techniques in algebraic multigrid (AMG) in order to improve load balances in the communication, thereby increasing the benefits of asynchrony.  We give an overview of these concepts and show several examples in support of the approach.
===affil3:
University of Illinois at Urbana-Champaign
===title:
Hiding Communication Costs in Sparse Matrix-Vector Multiplication and Algebraic Multigrid
===affil2:
University of Illinois at Urbana-Champaign
===lastname2:
Olson
===firstname4:
William D.
===keyword1:
Optimization of Complex Problems/Systems (PDE-constrained optimization, optimization under uncertainty, and simulation-based optimization)>Optimization of Complex Problems/Systems
===workshop:
no
===lastname:
Bienz
===firstname5:

===keyword2:
NOT_SPECIFIED
===otherauths:

===affil4:
University of Illinois at Urbana-Champaign
===competition:
no
===firstname7:

===firstname6:

===keyword_other1:
Parallel Scalability of Sparse Matrix-Vector Multiplication and Algebraic Multigrid
===lastname5:

===affilother:

===firstname2:
Luke