QoS-Aware Fog Computing Resource Allocation Using Feasibility-Finding Benders Decomposition

Abstract

We investigate a joint offloading and resource allocation under a multi-layer cooperative fog and cloud computing architecture, aiming to minimize the total energy consumption of mobile devices while meeting users' QoS requirements, e.g., delay, security, and application compatibility. Due to the mutual coupling amongst offloading decision and resource allocation variables, the resulting optimization is a mixed integer nonlinear programming problem that is NP-hard. Such problem often requires exponential time to find the optimal solution. In this work, we propose a distributed approach, namely feasibilityfinding Benders decomposition (FFBD), that decomposes the original problem into a master problem for the offloading decision and subproblems for resource allocation. These (simpler) subproblems can be solved in parallel at fog nodes, thereby reducing both the complexity and the computational time. The numerical results show that the FFBD always returns the optimal solution of the problem with significantly less computation time (e.g., in comparing with the branch-and-bound method).

Survey on Placement Methods in the Edge and Beyond

“…In terms of subjects to optimization, the offloading decisions are often complemented with additional aspects, other than the aforementioned positioning of mobile edge nodes. Therefore, next to the research results about offloading decision optimization, e.g., [84], [94], [115], [148], [157], [160], [183], there are several works that propose to jointly optimize offloading and compute resource allocation in the edge infrastructure, e.g., [101], [102], [112], [113]. In these papers the authors strive to optimize the operation of the overall MEC system while end users and/or terminals aim at minimizing their own battery usage or the processing time that constitutes the service latency.…”

“…In the MEC offloading papers the goal is selected from a rather limited set of inherent choices: energy and processing time. One can find papers in which one of these two goals are set out, e.g., energy in [84], [85], [102], processing time in [94], [104], [115], [148], [152], [157], [173], and there are also related works in which the goals are targeted jointly, e.g., in [81], [103], [112], [113], [117], [183]. While the former goal aims at preserving the limited battery capacity of terminals, e.g., IoT sensors, mobile phones, [57], [60], [63], [69], [74], [78], [79], [81], [82], [84]- [87], [90], [93], [94], [98], [101]- [104], [112], [113], [115], [117], [130], [143], [144], [148], [151], [152], [154], [157]-…”

“…The remaining dimension of the second group in our taxonomy, Security and privacy appear in [102], [151], [183]. The authors of [183] points out that two potential privacy issues are induced by the wireless task offloading feature of MEC: location privacy and usage pattern privacy.…”

“…The edge devices buy resources from their immediate neighbors, if their tasks are to be considered private. In [102] the authors consider different user-defined QoS requirements in terms of delay, compatibility and security. They apply Bender decomposition techniques to decompose the original offloading problem into a master problem and subproblems that can be solved in parallel at fog nodes.…”

“…Convex optimization [56], [112] Convex and quasi-convex optimization [101] Geometric programming [60] Interior-point method Linear programming [60], [102], [116] Benders decomposition [86] LP with fractal solution rounding Stochastic control [108] Sine cosine algorithm Other heuristic solution [68], [81], [93], [112], [139], [154] n/a [113]…”