Fair multi-resource allocation with external resource for mobile edge computing

Erfan Meskar, Ben Liang

Abstract: We consider the problem of fair multi-resource allocation for mobile edge computing (MEC). In MEC, to execute tasks with demands on multiple types of computing resources in the edge servers, the users must upload their tasks over a single dedicated wireless communication link that exists outside the servers. For this environment, we design a multi-resource allocation mechanism that extends the notion of dominant resource fairness (DRF) to accommodate an external resource, called DRF-ER. It provides several highly desirable properties. First, DRF-ER is envy-free, as no user prefers the allocation of another user. Second, DRF-ER allocations are Pareto optimal, as no one can improve its allocation without decreasing that of the others. Finally, DRF-ER is strategy-proof, as no user has an incentive to lie about its resource demand. Large-scale simulation driven by Google cluster traces further shows that DRF-ER significantly outperforms a naive extension of DRFH, which is a well-known variant of DRF for multiple servers, leading to higher resource utilization.