Journal of KIISE:Computing Practices and Letters (한국정보과학회논문지:컴퓨팅의 실제 및 레터)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

An Analysis and Comparison on Efficiency of Load Distribution Algorithm in a Clustered System

클러스터 시스템의 부하분산 알고리즘의 효율성 비교분석

  • 김석찬 (계명대학교 산업시스템공학과) ;
  • 이영 (계명대학교 산업시스템공학과)
  • Published : 2006.04.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this thesis, we analyze the efficiency of the algorithm to distribute the load in the clustered system, by comparing with the existed algorithm. PWLC algorithm detects each server's load in the system at weighted period, and following the detection of the loads, a set of weights is given to each server. The system allocates new loads to each server according to its weight. PWLC algorithm is compared with DWRR algorithm in terms of variance, waiting time by varying weighted Period. When the weighted period is too short, the system bears a heavy load for detecting load over time. On the other hand, when the weighted period is too long, the load balancing control of the system becomes ineffective. The analysis shows PWLC algorithm is more efficient than DWRR algorithm for the variance and waiting time.

본 연구에서는 클러스터 시스템에 적용되는 새로운 부하할당 알고리즘을 기존의 알고리즘과 비교하여 분석하고자 한다. PWLC 알고리즘은 설정된 가중치 산정주기마다. 시스템의 부하를 감지하여, 각 서버에 가중치를 부여하여 다음 주기에 가중치에 의하여 부하를 분산시키는 알고리즘이다. PWLC 알고리즘과 DWRR 알고리즘을 가중치 산정주기를 변화시키면서 분산과 대기시간 등에 비교하였다. 가중치 산정주기가 너무 짧으면 시스템은 부하를 감지하는데 잉여부하가 소요될 수 있으며, 이와 반대로, 가중치 산정주기가 너무 길면 알고리즘 적용에 의한 부하할당이 비효율적으로 될 수 있다. PWLC 알고리즘이 DWRR 알고리즘보다. 더 효율적임을 알 수 있다.

Keywords

References

  1. Patrick Killelea, Web Performance Tuning, O'Reilly 2 edition October, 2001
  2. IBM corp, IBM Technical Report CS 260, www. ibm.com, 2004
  3. Der-Chiang Li, Chihsen Wu, Fengming M. Chang, Determination of the parameters in the dynamic weighted Round-Robin method for network load balancing, computer & operations research, Vol. 32, pp. 2129-2145, 2005 https://doi.org/10.1016/j.cor.2004.02.004
  4. Wensong zhang, Linux Virtual Seroer for Scalable Network Services, Ottawa Linux Symposium 2000
  5. Gregory F. Pfister, In Search of Clusters Second Edition, Prentice-Hall PTR, New Jersey, 1998
  6. microsoft corporation, Server Cluster Architecture, http://www.microsoft.com/windows.netserver/techinfo/overview/servercluster.mspx, 2002
  7. Valeraia C, Michele c, Philip S. Yu, Dynamic load balancing on web-server systems, IEEE Internet Computing, pp. 28-39, May, 1999 https://doi.org/10.1109/4236.769420
  8. Shan Hong Li, Sung Ki Kim, Yong hee Na, Byoung joon Min, Analysis of the Load Balancing Algorithms according to the Request Pattern on the LVS Cluster Systems, Korean Information Proceeding Society, Vol. 9-A No.2, 10, 2002
  9. Sun Young Park, Do Hyun Park, jun Won Lee, Jeong Wan jo, Back-end Prefetching Scheme for Improving the Performance of Cluster-based Web Servers, Korean Information Proceeding Society, Vol. 29-A, No.5, pp. 265-273, 6. 2002
  10. Jin Young Kim, Seong Cheon Kim, Effective Prioritized HRW Mapping in Heterogeneous Web Server Cluster, Korean Information Proceeding Society, Vol. 30-A, No. 12, pp. 708-713, Dec 2003
  11. Y. Rhee, S.C Kim, A study of distributing the load of the LVS clustering system based on the dynamic weight, Korean Information Proceeding Society, Vol. 8-A, No.4, pp. 299-310, Dec 2001
  12. Y. Rhee, A Study of the load distributing algorithm on the heterogeneously clustered web system, Korean Information Proceeding Society, Vol. 10-A, No.3, pp. 225-230, Aug. 2003 https://doi.org/10.3745/KIPSTA.2003.10A.3.225
  13. Y. Rhee, S.C Kim, System Infrastructure of Efficient Web Cluster System to decrease the Response Time using the Load Distribution Algorithm, Korean Information Science Society, Vol. 10-6, pp. 507-513, Dec 2004
  14. Nadimpalli S, Manumdar S. Techniques for achieving high performance Web Servers. International conference on IEEE Parallel Processing, p. 233-241, 2000 https://doi.org/10.1109/ICPP.2000.876130