Estimated system reliability of a cloud computing network subject to maintenance budget

Abstract

This paper proposes an estimation method to evaluate the capability of a cloud computing network. To keep a good quality of service, maintenance actions are necessary when the CCN falls to a specific state such that it cannot afford enough capacity to meet d units of data. Another crucial factor to consider is the transmission time for sending data from the cloud to the client. Given the maintenance budget B and within time T, we estimate the system reliability that is defined as the probability that d units of data can be sent from the cloud to the client under both B and T. Several disjoint minimal paths are utilized to send data simultaneously such that the transmission time is decreased. To estimate the system reliability, a bounding approach is proposed to generate two sets of capacity vectors, UB-MPs and LB-MPs, where a UB-MP is the minimal capacity vector satisfying d and T while an LB-MP is the minimal capacity vector satisfying d, B, and T. Subsequently, the upper and lower bounds of system reliability can be computed in terms of such vectors by applying the recursive sum of disjoint products algorithm.