Energy-Aware Application Placement in Mobile Edge Computing: A Stochastic Optimization Approach

The Quality of Service (QoS) in Mobile Edge Computing (MEC) systems is significantly dependent on the application offloading and placement decisions. Due to the movement of users in MEC networks, an optimal application placement might turn into the least efficient placement in few minutes. Thus, it...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on parallel and distributed systems 2020-04, Vol.31 (4), p.909-922
Main Authors: Badri, Hossein, Bahreini, Tayebeh, Grosu, Daniel, Yang, Kai
Format: Article
Language:English
Subjects:
Algorithms
Cloud computing
Computational modeling
Edge computing
Energy budget
Energy consumption
Energy management
energy-aware application placement
Mobile computing
Mobile edge computing
multi-stage stochastic programming
Optimization
parallel algorithms
Placement
Quality of service
sample average approximation
Servers
Stochastic processes
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The Quality of Service (QoS) in Mobile Edge Computing (MEC) systems is significantly dependent on the application offloading and placement decisions. Due to the movement of users in MEC networks, an optimal application placement might turn into the least efficient placement in few minutes. Thus, it is crucial to take the dynamics of the system into account when designing application placement mechanisms. On the other hand, energy consumption of servers is a significant component of the cost of services in MEC systems and must also be considered in the design of the mechanisms. In this article, we model the problem of energy-aware application placement in edge computing systems as a multi-stage stochastic program. The objective is to maximize the QoS of the system while taking into account the limited energy budget of the edge servers. To solve the problem, we design a novel parallel Sample Average Approximation (SAA) algorithm. We conduct an extensive experimental analysis to evaluate the performance of the proposed algorithm using real-world trace data.
ISSN:1045-9219
1558-2183
DOI:10.1109/TPDS.2019.2950937