An Intelligent Reservation Algorithm for Workload Distribution

부하 분산을 위한 지능형 예약 알고리즘

  • Published : 2005.08.01

Abstract

We proposes an algorithm of measuring computation loads to distribute workload of clients. The key of the algorithm is to transfer a suitable amount of processing demand from senders to receivers in relocating intrinsically. This amount is determined dynamically during sender-receiver negotiations. Factors considered when this amount is determined include processing speeds of different nodes, the current load state of both sender and receiver, and the processing demands of tasks eligible for relocation. We also propose a load state measurement scheme which is designed particularly for heterogeneous systems. Based on this analysis, we design new algorithm for supporting heterogeneous distributed web server system, and compare their performance against other existing algorithms. We show that the new algorithm improve the performance of CPU utilization and response time.

본 논문에서는 클라이언트들의 작업부하를 분산하기 위한 측정 알고리즘을 제안하였다. 이 알고리즘은 작업들이 재배치되는 것을 허용할 때, 송신자가 수신자에게 요청한 처리량만큼 적절하게 전송하여야 한다. 이 양은 송신자와 수신자의 협상 과정에서 동적으로 결정된다. 전송하여야 할 작업량을 결정할 때 전송 노드는 수신하는 다른 노드들의 처리 속도를 포함하여 송신 노드와 수신 노드의 현재 부하상태, 재배치하기에 적절한 작업의 처리 요구량 등의 요소들을 고려하여야 한다. 이러한 분석에 근거하여 이기종 분산 웹서버 시스템에 적합한 새로운 알고리즘을 제안하고, 기존 알고리즘과의 성능을 비교하였다. 시뮬레이션을 통하여 본 논문에서 제안한 알고리즘이 CPU 이용율과 응답시간에서 성능의 개선을 보였다.

Keywords

References

  1. Agarwal, P., MTech Thesis, CSE, IIT Kanpur. 'A testbed for performance evaluation of load balancing strategies of web server system', May, 2001
  2. Cisco Systems Inc. 'Distributed Director White Paper'. http://www.cisco.com/warp/public/cc/cisco/mkt/scale/distr/tech/d_wp.htm
  3. 'Apache Keep-Alive support.' This document can be obtained from http://www.apache.org/docs-1.2/keepalive.html
  4. Daemon9. Libnet: Network Routing Library, Aug. 1999 'This document can be obtained from http://www.packet/actory.net/libnet/
  5. Mosberger, D., and Jin, T. httperf-A Tool for Measuring Web Server Performance. This document can be obtained from ftp://ftp.hpl.hp.com/pub/httperf/
  6. V. Cardellini, M. Colajanni, and P. S. Yu. Dynamic load balancing on Web-server systems. IEEE Internet Computing, 3(3):pp.28-39, May 1999 https://doi.org/10.1109/4236.769420
  7. M. F. Arlitt, R. Friendrich, and T. Jin. Workload characterization of a Web proxy in a cable modem environment. ACM Performance Evaluation Review, 27(2):pp.25-36, Aug. 1999 https://doi.org/10.1145/332944.332951
  8. J. E. Pitkow. Summary of WWW characterizations. World Wide Web, 2(1-2):pp.3-13, Mar. 1999 https://doi.org/10.1023/A:1019284202914
  9. N. G. Shibaratri, and P. Krueger. Two adaptive location policies for global scheduling algorithms. In Proceedings, The 10th International Conference on Distributed Computing Systems, pp. 502-509, May 1990
  10. T. Schroeder, S. Goddard, and B. Ramamurthy. Scalable Web server clustering technologies. IEEE Network, pp.38-45, May 2000
  11. 이준연, 임재현, '작업 이주시 보장/예약 기법을 이용한 프로세서 쓰레싱 빈도 감소', 정보처리학회논문지A, 8(2):pp.133-146, Jun. 2001