Files in this item

FilesDescriptionFormat

application/pdf

application/pdfspmm_tr.pdf (1MB)
(no description provided)PDF

Description

Title:Optimizing Sparse Matrix-Matrix Multiplication for the GPU
Author(s):Dalton, Steven; Bell, Nathan; Olson, Luke
Subject(s):parallel
sparse
gpu
matrix-matrix
Abstract:Sparse matrix-matrix multiplication (SpMM) is a key operation in numerous ar- eas from information to the physical sciences. Implementing SpMM efficiently on throughput-oriented processors, such as the graphics processing unit (GPU), requires the programmer to expose substantial fine-grained parallelism while conserving the limited off-chip memory bandwidth. Balancing these concerns, we decompose the SpMM operation into three, highly-parallel phases: expansion, sorting, and compres- sion, and introduce a set of complementary bandwidth-saving performance optimiza- tions. Our implementation is fully general and our optimizations lead to substantial efficiencies for a SpMM product.
Issue Date:2013-03-26
Genre:Technical Report
Article
Type:Text
Language:English
URI:http://hdl.handle.net/2142/42667
Publication Status:published or submitted for publication
Peer Reviewed:is peer reviewed
Date Available in IDEALS:2013-03-26


This item appears in the following Collection(s)

Item Statistics