Service Placement in Fog Computing Using Constraint Programming

Abstract

This paper investigates whether the use of Constraint Programming (CP) could enable the development of a generic and easy-to-upgrade placement service for Fog Computing. Our contribution is a new formulation of the placement problem, an implementation of this model leveraging Chocosolver and an evaluation of its scalability in comparison to recent placement algorithms. To the best of our knowledge, our study is the first one to evaluate the relevance of CP approaches in comparison to heuristic ones in this context. CP interleaves inference and systematic exploration to search for solutions, letting users on what matters: the problem description. Thus, our service placement model not only can be easily enhanced (deployment constraints/objectives) but also shows a competitive tradeoff between resolution times and solutions quality.

An Overview of Service Placement Problem in Fog and Edge Computing

“…It uses a set of constraints that can easily be extended further to involve more aspects. For instance, in [9], Ait-Salaht et al propose to handle the SPP problem by provided a generic and easy-to-upgrade constraint programming model. Brogi et al [29,31] propose a constrained model to determine the feasible deployments (if any) of an application in the Fog infrastructure.…”

“…C/Off/S/nM Exact [9] Provides feasible (resp. optimal) service placement solutions in Fog environment.…”

Latency-aware Resource Management at the Edge

“…Ahvar [67] Aït-Salaht [85] Bittencourt [78] Borgia [86] Borylo [71] Boubin [87] Chen [81] De Maio [88] Elgazar [83] Fahs [89] Fricker [72] Gu [80] Habak [20] Liu [77] Liu [50] Mahmud [90] Penner [52] Rodrigues [73] Sardellitti [82] Singh [40] Skarlat [46] Sonmez [91] Tang [79] Tärneberg [75] Wang [45] Wang [70] Wang [92] Wang [93] Xia [94] Yi [76] Zamani [74] 3.4. Objective user.…”

“…As task placement is one part of the orchestration framework presented in the thesis, a further litterature study showed that in the majority of the works, a central entity gets all the task placement requests. In this type of scenario, the task placement problem can be formulated as an optimization problem where several tasks have to be dispatched among several edge devices and is solved using various optimization techniques [95,70,96,97,85]. Even when the requests are received in a distributed manner among of group of edge devices, the task placement is still performed at only one device within the group [98].…”

“…Other works use mixed-integer non-linear programming [80,76]. Recently, constraint programming [85] has also been used.…”