Evaluation of the performance of parallel sparse-matrix multiplication and the effect of dynamic load-balancing

被引：0

作者：

Nanri, Takeshi ^{[1
]}

Soga, Takeshi ^{[2
]}

Kurihara, Koji ^{[3
]}

Gu, Feng Long ^{[1
]}

Ishihata, Hiroaki ^{[4
]}

Murakami, Kazuaki ^{[1
,3
]}

机构：

[1] Kyushu Univ, Res Inst Informat Technol, 6-10-1 Hakozaki Fukuoka, Fukuoka 8128581, Japan

[2] IST, Fukuoka, Japan

[3] Kyushu Univ, Grad Sch Informat Sci & Elect Engn, Fukuoka 812, Japan

[4] Fujitsu Ltd, Fukuoka, Japan

来源：

COMPUTATION IN MODERN SCIENCE AND ENGINEERING VOL 2, PTS A AND B | 2007年 / 2卷

关键词：

sparse matrix; parallel computing; load balance; matrix multiplication; MPI;

D O I：

暂无

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper proposes and evaluates an algorithm of parallel matrix multiplication on compressed sparse-matrices. In addition to that, to overcome the problem of imbalanced distribution of non-zero elements over processors, a technology for dynamic load-balancing is examined. Measurements of the performance show that the algorithm achieves sufficient speed up. The experiments in this paper could not show the significant effect of the dynamic load-balancing. It requires more analysis to determine the reason of this.

引用

页码：106 / +

页数：2

共 50 条

[31] A locality-conscious load-balancing based on negotiations in dynamic unstructured mesh computations [J].

Mohamed, AS ;

Baydogan, VS .

PDPTA'03: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED PROCESSING TECHNIQUES AND APPLICATIONS, VOLS 1-4, 2003, :806-813

[32] Novel Multi-Level Dynamic Traffic Load-Balancing Protocol for Data Center [J].

Memon, Sheeba ;

Huang, Jiawei ;

Saajid, Hussain ;

Bux, Naadiya Khuda ;

Saleem, Arshad ;

Aljeroudi, Yazan .

SYMMETRY-BASEL, 2019, 11 (02)

[33] Parallel cellular automata for large-scale urban simulation using load-balancing techniques [J].

Li, Xia ;

Zhang, Xiaohu ;

Yeh, Anthony ;

Liu, Xiaoping .

INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE, 2010, 24 (06) :803-820

[34] Reducing inter-process communication overhead in parallel sparse matrix-matrix multiplication [J].

Ahmed M.S. ;

Houser J. ;

Hoque M.A. ;

Raju R. ;

Pfeiffer P. .

International Journal of Grid and High Performance Computing, 2017, 9 (03) :46-59

[35] THE SCHEDULING OF SPARSE MATRIX-VECTOR MULTIPLICATION ON A MASSIVELY PARALLEL DAP COMPUTER [J].

ANDERSEN, J ;

MITRA, G ;

PARKINSON, D .

PARALLEL COMPUTING, 1992, 18 (06) :675-697

[36] Parallel dynamic load balancing strategies for adaptive irregular applications [J].

Biswas, R ;

Das, SK ;

Harvey, DJ ;

Oliker, L .

APPLIED MATHEMATICAL MODELLING, 2000, 25 (02) :109-122

[37] SIMULTANEOUS INPUT AND OUTPUT MATRIX PARTITIONING FOR OUTER-PRODUCT-PARALLEL SPARSE MATRIX-MATRIX MULTIPLICATION [J].

Akbudak, Kadir ;

Aykanat, Cevdet .

SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2014, 36 (05) :C568-C590

[38] Performance Evaluation of Matrix Multiplication in Virtual Machine [J].

Muhammad, Asif ;

Islam, Muhammad Arshad .

PROCEEDINGS OF 2017 INTERNATIONAL CONFERENCE ON COMMUNICATION, COMPUTING AND DIGITAL SYSTEMS (C-CODE), 2017, :205-210

[39] A High-Performance Parallelization and Load-Balancing Approach for Modern Power-Systems [J].

Khaitan, Siddhartha Kumar ;

McCalley, James D. .

INTERNATIONAL JOURNAL OF BUSINESS ANALYTICS, 2015, 2 (02) :62-74

[40] A generic strategy for dynamic load balancing of dynamic parallel distributed mesh generation [J].

Yuan, Youwei ;

Guo, Qingqing .

DCABES 2006 PROCEEDINGS, VOLS 1 AND 2, 2006, :199-202

← 1 2 3 4 5 →