Queueing theoretic approach to job assignment strategy considering various inter-arrival of job in fog computing
Tomoya Mori, Yoichi Utsunomiya, Xuejun Tian, Takashi Okuda · 2017
Recent years, various services using Cloud computing are involving in people's life. However, when techniques and devices of Internet of Things (IoT) are getting more involved in people's life, current Cloud computing can hardly satisfy the requirement of low latency. To address the above problem, new distributed computing called Fog computing is proposed by Cisco System, Inc. Fog computing has a simple three-tiered hierarchy: Edge device, Fog and Cloud. The Fog tier is put between the Edge device tier and the Cloud tier to play the same or part of roles of the Cloud tier by its proximity to Edge device. The Fog tier consists of some fog nodes. The fog node receives jobs from the Edge device tier. Then, according to the jobs, the fog node decides to either process the jobs or assign it Cloud tier immediately. We call this process “Job assignment”. However, the division of roles between the Fog tier and the cloud tier is not clear and it must be decided depending on the system. In this paper, we express the Fog computing with VCHS (Various Customers, Heterogeneous Servers) queueing model. We then evaluate this model and study on an effective Job assignment strategy.