,
HPCC'2012
Space-efficient sparse matrix storage formats for massively parallel systems
Author: I. Šimeček and D. Langr and P. Tvrdík
Keywords

data reduction;distributed file system;large sparse matrices;massively parallel computer systems;matrix structure information;parallel I-O system;processor cores;space complexity;space-efficient sparse matrix storage formats;computational complexity;distributed databases;multiprocessing systems;parallel processing;sparse matrices;


Abstract

In this paper, we propose and evaluate new storage formats for sparse matrices that minimize the space complexity of information about matrix structure. The motivation of our work are applications with very large sparse matrices that due to their size must be processed on massively parallel computer systems consisting of tens or hundreds of thousands of processor cores and that must be stored in a distributed file system using parallel I/O. The parallel I/O is typically the main performance bottleneck and reading or writing such matrices from/to distributed file system can take significant amount of time. We try to reduce this time by reducing the amount of data to be processed.

 



BibTex entry:
@INPROCEEDINGS{JA_HPCC12,
author = {I. Šimeček and D. Langr and P. Tvrdík},
title = {Space-efficient sparse matrix storage formats for massively parallel systems},
booktitle = {High Performance Computing and Communication and 2012 IEEE 9th International Conference on Embedded Software and Systems (HPCC-ICESS)},
year = {2012},
series = {HPCC'12},
pages = {54--60},
address = {Liverpool, Great Britain},
month = {june},
doi = {10.1109/HPCC.2012.18},
isbn = {978-0-7695-4749-7/12},
keywords = {data reduction;distributed file system;large sparse matrices;massively parallel computer systems;matrix structure information;parallel I-O system;processor cores;space complexity;space-efficient sparse matrix storage formats;computational complexity;distributed databases;multiprocessing systems;parallel processing;sparse matrices;},
unique-id = {WOS:000310377500008} }

BACK