Using a specialized Markov chain in the reliability model of disk arrays RAID-10 with data mirroring and striping

International Conference on Mechanical Engineering, Automation and Control Systems 2016

机械制造;无线电电子学;计算机科学

Rahman, P.A.^1

Department of Automated Technological and Informational Systems, Ufa State Petroleum Technological University, 2, October ave., Sterlitamak

453118, Russia^1

关键词: Data loss;    Disk array;    Fault tolerant systems;    Fault-tolerant;    Mean time of failures;    Reliability model;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/177/1/012087/pdf
DOI  :  10.1088/1757-899X/177/1/012087

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

This paper deals with a specialized Markov chain for a reliability model of the fault-tolerant system with dual-type failures, non-synchronized repairs and full restore after reaching the failed state. A generalized iterative procedure for calculation of the system's stationary availability factor, the mean time of repair and the mean time of failure is introduced. An application of the specialized Markov chain in the reliability model of the fault-tolerant dual-level disk arrays 'RAID-10' and calculation of the mean time to data loss by using the generalized iterative procedure are also observed. Finally, calculation examples of mean time to data loss of the RAID-10 array are also given.