Edge cloud resource expansion and shrinkage based on workload for minimizing the cost

摘要

The joint use of cloud and edge has been paid more and more attention by enterprises. Enterprises use the cloud's utility computing and elasticity to placement resources flexibly under the adaptive load. However, there are still many challenges in how to minimize edge cloud costs and ensure data integrity in the resource expansion and shrinkage of edge cloud. This paper studies the resource management problem based on load balance in edge and cloud environments. The novelty is that we consider the resource granularity of the cloud service provider and the data loss problem that resource shrinkage may bring. Firstly, the resource expansion and shrinkage model based on service cost is proposed to reduce the cost of edge cloud clusters under the condition of satisfying the cluster load. Secondly, due to the problem of unbalanced cluster load caused by resource expansion and the data reliability caused by resource shrinkage, a data migration model is proposed aiming at the data loss problem that resource shrinkage may bring. Finally, the proposed algorithms are evaluated with extensive comparisons among the benchmarks and sensitivity tests to various environment factors. (C) 2019 Elsevier B.V. All rights reserved.