Dynamic partner identification in mobile agent-based distributed job workflow execution

Yuhong Feng, Wentong Cai, Jiannong Cao

Research output: Journal article publicationJournal articleAcademic researchpeer-review

5 Citations (Scopus)

Abstract

This paper is concerned with the design, implementation, and evaluation of algorithms for communication partner identification in mobile agent-based distributed job workflow execution. We first describe a framework for distributed job workflow execution over the Grid: the Mobile Code Collaboration Framework (MCCF). Based on the study of agent communications during a job workflow execution on MCCF, we identify the unnecessary agent communications that degrade the system performance. Then, we design a novel subjob grouping algorithm for preprocessing the job workflow's static specification in MCCF. The obtained information is used in both static and dynamic algorithms to identify partners for agent communication. The mobile agent dynamic location and communication based on this approach is expected to reduce the agent communication overhead by removing unnecessary communication partners during the dynamic job workflow execution. The proof of the dynamic algorithm's correctness and effectiveness are elaborated. Finally, the algorithms are evaluated through a comparison study using simulated job workflows executed on a prototype implementation of the MCCF on a LAN environment and an emulated WAN setup. The results show the scalability and efficiency of the algorithms as well as the advantages of the dynamic algorithm over the static one.
Original languageEnglish
Pages (from-to)1137-1154
Number of pages18
JournalJournal of Parallel and Distributed Computing
Volume67
Issue number11
DOIs
Publication statusPublished - 1 Nov 2007

Keywords

  • Communication partner identification
  • Distributed job workflow execution
  • Grid computing
  • Mobile agent communication

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computer Networks and Communications
  • Artificial Intelligence

Cite this