An Availability Model for Active / Active Cluster Systems


The KIPS Transactions:PartC, Vol. 8, No. 2, pp. 173-181, Apr. 2001
10.3745/KIPSTC.2001.8.2.173,   PDF Download:

Abstract

For the fast increase of a hardware reliability and the size or complexity of application softwares, the ratio of hardware-origin system failures becomes much smaller than that of software-origin system failures. To improve the availability of an Active/Active cluster systems, we study software rejuvenation method that is a proactive fault-tolerant approach to handle software-origin system failures. In this paper, we model software rejuvenation and switchover processes as the problem of a semi-Markov chain and get mathematical steady-state solution of the process. We calculate the availability and the downtime of an Active/Active cluster systems using software rejuvenation method and find that the method can be improve the availability of the systems.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
K. J. Park and S. S. Kim, "An Availability Model for Active / Active Cluster Systems," The KIPS Transactions:PartC, vol. 8, no. 2, pp. 173-181, 2001. DOI: 10.3745/KIPSTC.2001.8.2.173.

[ACM Style]
Kie Jin Park and Sung Soo Kim. 2001. An Availability Model for Active / Active Cluster Systems. The KIPS Transactions:PartC, 8, 2, (2001), 173-181. DOI: 10.3745/KIPSTC.2001.8.2.173.