Evaluation Criteria for Sparse Matrix Storage Formats

When authors present new storage formats for sparse matrices, they usually focus mainly on a single evaluation criterion, which is the performance of sparse matrix-vector multiplication (SpMV) in FLOPS. Though such an evaluation is essential, it does not allow to directly compare the presented forma...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on parallel and distributed systems 2016-02, Vol.27 (2), p.428-440
Main Authors: Langr, Daniel, Tvrdik, Pavel
Format: Article
Language:English
Subjects:
Communities
Computer networks
Construction
Criteria
Evaluation criterion
Format
Indexes
Matrix converters
matrix-vector multiplication
memory footprint
Memory management
Multiplication
nonzero matrix structure
Runtime
Sparse matrices
sparse matrix
storage format
test matrices
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:When authors present new storage formats for sparse matrices, they usually focus mainly on a single evaluation criterion, which is the performance of sparse matrix-vector multiplication (SpMV) in FLOPS. Though such an evaluation is essential, it does not allow to directly compare the presented format with its competitors. Moreover, in case that matrices are within an HPC application constructed in different formats, this criterion alone is not sufficient for the key decision whether or not to convert them into the presented format for the SpMV-based application phase. We establish ten evaluation criteria for sparse matrix storage formats, discuss their advantages and disadvantages, and provide general suggestions for format authors/evaluators to make their work more valuable for the HPC community.
ISSN:1045-9219
1558-2183
DOI:10.1109/TPDS.2015.2401575