PROJECT TITLE :

Performability Modeling for RAID Storage Systems by Markov Regenerative Process - 2018

ABSTRACT:

This Project presents a performability model for RAID storage systems using Markov regenerative method to check different RAID architectures. Whereas homogeneous Markov models are extensively used for reliability analysis of RAID storage systems, the memory-less property of the sojourn time assumed in such models isn't satisfied actually, especially in disk rebuild process whose progress is not interrupted even at an event of another disk failure. During this Project, we have a tendency to use Markov regenerative method which allows us to model the widely distributed rebuild times providing a needed extension of the ancient Markov models. The Markov regenerative process is then used to assess the performability of the storage system by assigning reward rates to each state primarily based on the $64000 storage benchmark results. Our numerical study characterizes the performability advantage of RAID6 design over RAID10 architecture in terms of sequential browse access. Our findings embody that the effect of exponential assumption for the rebuild times has practically negligible effect when we concentrate on data availability. But, the effect this approximation on performability prediction might not be negligible particularly when the performance level drastically changes in degraded states. Our MRGP model provides additional correct prediction of performability in such cases.