A partitioning method of balancing CPU utilization of servers in DVE

  • Authors:
  • Dongkee Won;Tingting Li;Beobkyun Kim;Dongun An;Seungjong Chung

  • Affiliations:
  • Chonbuk National University;Chonbuk National University;Korea Institute of Science and Technology, Korea, TingTing Li China;Chonbuk National University;Chonbuk National University

  • Venue:
  • PDCN '08 Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Networks
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

In Distributed Virtual Environment (DVE), multiple users and multiple virtual environments are connected via internet to share the information and interact where they are in the different locations. Therefore, how to decrease the growing requirement of bandwidth and computational resources is one of the major challenges in the field of designing a scalable and cost-effective DVE. Actually, the partitioning problem is one of efficient issues on designing an excellent DVE. A brilliant partitioning method is related with assigning several avatars into the suitable servers with well balancing the growing requirement of bandwidth and computational resources in DVE. In this paper, a new method LCAA is proposed. The LCAA is a new partitioning method that balancing the CPU utilization of servers in DVE especially. The illumination of LCAA method is based on such phenomenon that while any of the servers CPU utilization reaches at 100 percentages, the DVE will present saturation state. In that case, a bad performance will be supported in DVE. In order to avoid such phenomenon, assigning the avatars into the servers with low CPU utilization is the main idea of LCAA partitioning method. Thus, we can easily reduce the possibility of occurrence of system saturation in DVE.