Skip to Main content Skip to Navigation
Theses

Algorithms for Tasks Offloading on Multiple Mobile Edge Servers

Abstract : Computation offloading is one of the most promising paradigm to overcome the lack of computational resources in mobile devices. Basically, it allows the execution of part orall of a mobile application in the cloud. The main objective is to reduce both execution time and energy consumption for the mobile terminals. Unfortunately, even if clouds have rich computing and storage resources, they are usually geographically far from mobile applications and may suffer from large delays, which is particularly problematic for mobile applications with small response time requirements. To reduce this long delay, one of the emerging approach is to push the cloud to the network edge. This proximity gives the opportunity to mobile users to offload their tasks to “local” cloud for processing. An Edge Cloud can be seen as small data center acting as a shadow image of larger data centers. This geographical proximity between mobile applications and edge cloud means that the access delay can be greatly reduced, but affects also higher throughput, improved responsiveness and better scalability. In this thesis, we focus on computation offloading in mobile environment (Mobile Edge Computing - MEC), composed of several edge servers. Our goal is to explore new and effective offloading strategies to improve applications performances in both execution time and energy consumption, while ensuring application requirements. Our first contribution is a new offloading strategy in the case of multiple edge servers. Thenwe extend this strategy to include the Cloud. Both strategies have been evaluated theoretically and experimentally by the implementation of an offloading middleware. Finally, we propose a new elastic approach in the case of multitasking applications characterized by a graph of dependencies between tasks.
Complete list of metadata

https://tel.archives-ouvertes.fr/tel-03227418
Contributor : Abes Star :  Contact
Submitted on : Monday, May 17, 2021 - 11:51:09 AM
Last modification on : Wednesday, May 19, 2021 - 3:23:21 AM
Long-term archiving on: : Wednesday, August 18, 2021 - 6:26:43 PM

File

edgalilee_th_2019_mazouzi.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-03227418, version 1

Citation

Houssemeddine Mazouzi. Algorithms for Tasks Offloading on Multiple Mobile Edge Servers. Mobile Computing. Université Paris-Nord - Paris XIII, 2019. English. ⟨NNT : 2019PA131076⟩. ⟨tel-03227418⟩

Share

Metrics

Record views

66

Files downloads

74