Design and performance analysis of a message scheduling scheme for WLAN-based cluster computing

Authors:
Junghoon Lee;Mikyung Kang;Euiyoung Kang;Gyungleen Park;Hanil Kim;Cheolmin Kim;Seongbaeg Kim;Jiman Hong
Affiliations:
Dept. of Computer Science and Statistics, Cheju National Univ;Dept. of Computer Science and Statistics, Cheju National Univ;Dept. of Computer Education, Cheju National Univ;Dept. of Computer Science and Statistics, Cheju National Univ;Dept. of Computer Education, Cheju National Univ;Dept. of Computer Education, Cheju National Univ;Dept. of Computer Education, Cheju National Univ;School of Computer Science and Engineering, Kwangwoon Univ., Jeju Do, Republic of Korea
Venue:
ICCSA'06 Proceedings of the 2006 international conference on Computational Science and Its Applications - Volume Part IV
Year:
2006

Citing 7
Cited 0

A unified wireless LAN architecture for real-time and non-real-time communication services

IEEE/ACM Transactions on Networking (TON)
Fault Tolerant MPI for the HARNESS Meta-computing System

ICCS '01 Proceedings of the International Conference on Computational Sciences-Part I
Beyond the "Device as Portal": Meeting the Requirements of Wireless and Mobile Devices in the Legion Grid Computing System

IPDPS '02 Proceedings of the 16th International Parallel and Distributed Processing Symposium
Solving Engineering Applications with LAMGAC over MPI-2

Proceedings of the 9th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Guest Editors' Introduction: Wireless Grids--Distributed Resource Sharing by Mobile, Nomadic, and Fixed Devices

IEEE Internet Computing
An efficient bandwidth management scheme for a hard real-time fuzzy control system based on the wireless LAN

RSFDGrC'05 Proceedings of the 10th international conference on Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing - Volume Part II
Error modeling schemes for fading channels in wireless communications: A survey

IEEE Communications Surveys & Tutorials

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes and analyzes the performance of a task scheduling scheme that couples network schedule for master-slave style parallel computing cluster built on top of wireless local area networks. When a transmission fails, the proposed scheme selects another data and destination that can replace the subtask scheduled on the unreachable node with minimal cost, rather than hopelessly retransmits on the bad channel. Simulation results performed via ns-2 event scheduler, show that the proposed scheme minimizes the task migration, improves the computation time by maximally 35.4 %, increases the probability of successful retransmission, and finally survives the network failure as long as at least one node is reachable at each instance of time.