Locality sensitive request distribution for fog and cloud servers

Fog computing is meant to bring the cloud resource closer to the edge of the Internet so that devices can access the back end services much faster. Additionally, the services hosted at the fogs can be customized to fit the local needs. Because fogs are dispersed throughout the network, each installation will have limited resources. This makes resource management a very critical issue. In this paper, we present a two-step resource management approach for fog computing. The first step decides the allocation of the devices to the fogs. The fogs are allocated in a two-tiered manner. That is, for each device, a home fog and a pool of backup fogs are allocated. In the second step, the requests from the devices are distributed to the allocated fogs or the cloud. Using simulation studies, we compared the performance of the proposed request distribution algorithm against existing ones. The results indicate that our request distribution algorithm outperforms existing ones.