Retrial Queue for Cloud Systems with Separated Processing and Storage Units

TLDR: This paper considers a retrial queueing model for cloud computing systems where the processing unit (server) and the storage unit (buffer) are separated and presents a recursive scheme for computing the stationary probability of all the states.

Abstract: This paper considers a retrial queueing model for cloud computing systems where the processing unit (server) and the storage unit (buffer) are separated. Jobs that cannot occupy the server upon arrival are stored in the buffer from which they are sent to the server after some random time. After completing a service the server stays idle for a while waiting for either a new job or a job from the buffer. After the idle period, the server starts searching for a job from the buffer. We assume that the search time cannot be disregarded during which the server cannot serve a job. We model this system using a retrial queue with search for customers from the orbit and obtain an explicit solution in terms of partial generating functions. We present a recursive scheme for computing the stationary probability of all the states.