Performance Analysis of an Edge Computing SaaS System for Mobile Users

Nowadays, the ever increasing demand of mobile computing applications has triggered the shift from the centralized mobile Cloud Computing toward mobile Edge Computing (MEC). The main feature of MEC is to move computing and storage to the network edges (e.g., radio access point/base stations), enabling resource-limited mobile devices to support computation-intensive and latency-critical applications. This paper deals with the performance evaluation of a MEC system providing computational capabilities to users within a limited area, according to the software as a service (SaaS) paradigm. In particular, a Markov multi-server queuing system model with requests reneging is proposed in order to accomplish the performance evaluation of the MEC system and derive the minimum number of processors to be allocated in order to fulfill specific service requirements in terms of resulting requests dropping probability. Finally, the pertinence of the proposed Markov queueing model is validated by comparing the obtained analytical predictions with numerical results derived by resorting to extensive computer simulations carried out under the assumption of realistic operating conditions.