English  |  正體中文  |  简体中文  |  Items with full text/Total items : 6487/11649
Visitors : 28481667      Online Users : 116
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: http://ir.ncue.edu.tw/ir/handle/987654321/11896

Title: Parallelizing CLIPS-based Expert Systems by the Permutation Feature of Pattern Matching
Authors: Wu, Chao-Chin;Lai, Lien-Fu;Chang, Yu-Shuo
Contributors: 資訊工程學系
Keywords: Parallel computing;CLIPS;Cluster system;Expert system;Rule based
Date: 2010-03
Issue Date: 2012-07-02T02:27:15Z
Publisher: IEEE
Abstract: CLIPS is a non-algorithmic language designed especially for developing expert systems. To address the problem that CLIPS suffers from long execution time because of the characteristics of rule-based language, previously we have proposed a Grid-enabled parallel CLIPS language and a
dynamic load balancing programming model that can parallelize the execution of a CLIPS program automatically if the data can be inferred independently. In this paper, we investigate how to apply the idea of automatic parallelization to other kinds of applications. For instance, a rule usually requires choosing multiple data items from the knowledge base to match with. This kind of matching is a permutation problem. All the different permutations must be divided into partitions and assigned to slaves for independent inferences. A programmer only needs to use three simple directives to provide necessary information to automatically parallelize the execution of an application. Experiment results show that the
best speedup is 10.38 when executing a knowledge management system in a heterogeneous cluster system with 12 processor cores.
Relation: The 2nd International Conference on Computer Engineering and Applications (ICCEA 2010), Mar. 19-21, 2010: 214-218
Appears in Collections:[資訊工程學系] 會議論文

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