National Changhua University of Education Institutional Repository : Item 987654321/11880
English  |  正體中文  |  简体中文  |  Items with full text/Total items : 6491/11663
Visitors : 24928828      Online Users : 67
RC Version 3.2 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search

Please use this identifier to cite or link to this item:

Title: Using MPI to Execute a FuzzyCLIPS Application in Parallel in Heterogeneous Computing Systems
Authors: Wu, Chao-Chin;Lai, Lien-Fu;Chang, Yu-Shuo
Contributors: 資訊工程學系
Date: 2008-07
Issue Date: 2012-07-02T02:25:43Z
Publisher: IEEE
Abstract: FuzzyCLIPS is a knowledge-base programming language designed especially for developing fuzzy expert systems. However, it usually requires much longer execution time than algorithmic languages. To address this problem, we propose to design a parallel version of FuzzyCLIPS to efficiently utilize the computing resources in emerging cluster and grid systems. MPI, the de-facto standard for parallel programming, is used to provide the facilities of parallelization and message passing. In contrast with previous researches using MPMD mode, we adopt SPMD programming model to ease the exploiting of data parallelism. Furthermore, to adapt to the heterogeneous computing resources, the FuzzyCLIPS inference engine is extended with a built-in function of load balancing to allocate appropriate amount of data to each process at the rum time. We have implemented a human resources Web site to evaluate the performance of the proposed parallel FuzzyCLIPS. The results show that we can garner superlinear speedup and provide a more reasonable response time.
Relation: The 8th IEEE International Conference on Computer and Information Technology (IEEE CIT'2008). Sydney, Australia, Jul. 8-11, 2008: 279-284
Appears in Collections:[Department and Graduate Institute of Computer Science and Information Engineering] Proceedings

Files in This Item:

File SizeFormat

All items in NCUEIR are protected by copyright, with all rights reserved.


DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback