TY - GEN
T1 - Multi-Objective Resource Optimization for Hierarchical Mobile Edge Computing
AU - Yaqub, Umair
AU - Sorour, Sameh
N1 - KAUST Repository Item: Exported on 2022-06-27
Acknowledged KAUST grant number(s): OSR-2016-KKI-2899
Acknowledgements: The authors would like to acknowledge the support of this work by both KAUST's Office of Sponsored Research under Award No. OSR-2016-KKI-2899, and KFUPM's Deanship of Scientific Research through projects EE002355 and KAUST-002.
This publication acknowledges KAUST support, but has no KAUST affiliated authors.
PY - 2018
Y1 - 2018
N2 - Mobile edge computing (MEC) enables the computation of complex tasks efficiently on end-devices. It has been shown that MEC can potentially reduce the delay and offer energy-efficient operation. So far, the research has been focused on or limited to multi-server optimization for partial offloading or competing to offload to the best server among different sets of mobile device. In this paper, we propose a hierarchical structure where user devices that have available resources can act as servers without the need to offload to the cloud. Results show that this model improves the system utility. Furthermore, there is a trade-off between the number of users in the network and the number of peers willing to accept tasks. As the number of users increase, the utility of the hierarchical system keeps increasing whereas the simple MEC system utility saturates.
AB - Mobile edge computing (MEC) enables the computation of complex tasks efficiently on end-devices. It has been shown that MEC can potentially reduce the delay and offer energy-efficient operation. So far, the research has been focused on or limited to multi-server optimization for partial offloading or competing to offload to the best server among different sets of mobile device. In this paper, we propose a hierarchical structure where user devices that have available resources can act as servers without the need to offload to the cloud. Results show that this model improves the system utility. Furthermore, there is a trade-off between the number of users in the network and the number of peers willing to accept tasks. As the number of users increase, the utility of the hierarchical system keeps increasing whereas the simple MEC system utility saturates.
UR - http://hdl.handle.net/10754/679369
UR - https://ieeexplore.ieee.org/document/8648109/
UR - http://www.scopus.com/inward/record.url?scp=85063514607&partnerID=8YFLogxK
U2 - 10.1109/glocom.2018.8648109
DO - 10.1109/glocom.2018.8648109
M3 - Conference contribution
SN - 9781538647271
BT - 2018 IEEE Global Communications Conference (GLOBECOM)
PB - IEEE
ER -