Presentation 2013-11-08
The Approximate String Matching on the Hierarchical Memory Machine, with Performance Evaluation
MANDUHU M, Koji NAKANO, Yasuaki ITO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The Hierarchical Memory Machine (HMM) is a theoretical parallel computing model that captures the essence of computing on CUDA-enabled CPUs. The approximate string matching (ASM) for two strings X and Y of length ra and n is a task to find a substring of Y most similar to X. The main contribution of this paper is to show an optimal parallel algorithm for the approximate string matching on the HMM and to implement it on a CUDA-enabled CPU. Our algorithm runs in O(n/w+(mn)/(dw)+(nL)/p+(mnl)/p) on the HMM with d streaming processors, memory band width w, global memory access latency L, and shared memory access latency l. Further, we implement our algorithm on GeForce GTX 580 GPU and evaluate the performance. The experimental results show that the ASM of two strings of 1024 and 4M (= 2^<22>) characters can be computed in 419.6ms, while the sequential algorithm can compute it in 27720ms. Thus, our implementation on the GPU attains a speedup factor of 66.1 over the single CPU implementation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) memory machine models / approximate string matching / edit distance / GPU / CUDA
Paper # CPSY2013-41
Date of Issue

Conference Information
Committee CPSY
Conference Date 2013/11/1(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Computer Systems (CPSY)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) The Approximate String Matching on the Hierarchical Memory Machine, with Performance Evaluation
Sub Title (in English)
Keyword(1) memory machine models
Keyword(2) approximate string matching
Keyword(3) edit distance
Keyword(4) GPU
Keyword(5) CUDA
1st Author's Name MANDUHU M
1st Author's Affiliation Department of Information Engineering, Hiroshima University()
2nd Author's Name Koji NAKANO
2nd Author's Affiliation Department of Information Engineering, Hiroshima University
3rd Author's Name Yasuaki ITO
3rd Author's Affiliation Department of Information Engineering, Hiroshima University
Date 2013-11-08
Paper # CPSY2013-41
Volume (vol) vol.113
Number (no) 282
Page pp.pp.-
#Pages 6
Date of Issue