.png){kind=icon}
Return
Energy-efficient task scheduling for mobile edge computing with virtual machine I/O interference
.png){kind=icon}
PRE
AI
DOI：10.1016/j.future.2023.06.020.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Mobile edge computing (MEC) is expected to support the computation-intensive and delay-sensitive applications of mobile internet users. In this paper, we investigate the resource allocation of MEC with the effect of I/O interference among parallel virtual machines (VMs) while satisfying the quality of service (QoS) of tasks. Different from existing works, we propose a flexible task scheduling approach that combining parallel and sequential computing to minimize the computing energy consumption of MEC server. We formulate the task scheduling problem as a mixed-integer nonlinear programming (MINLP) and decompose it as a CPU resource allocation subproblem, a computing time slot subproblem, and a VM selection subproblem. We show the first subproblem is a convex problem and propose a CPU frequency allocation (CFA) algorithm based on the Karush-Kuhn-Tucker (KKT) conditions to obtain the optimal CPU frequency resource allocation. For the time slot allocation and VM selection subproblems, we propose the three step allocation (TSA) and urgency based adjusting (UBA) algorithms to obtain the near-optimal solutions, respectively. Simulation results show that compared with several time slot allocations and VM selections, the proposed TSA and UBA algorithms can save up to 21.7% and 95.8% of energy consumption, respectively.& COPY; 2023 Elsevier B.V. All rights reserved.
Keywords:
Mobile edge computing
Virtual machine
I
O interference
Task scheduling
Mixed-integer nonlinear programming
Karush-Kuhn-Tucker
Journal
F
IF:
6.1
Papers:
6.9K
Citations:
2.3W
Organization
Cited Papers
ShareSaveShareSaveShareSavesiRNA‐mediated GABABreceptor at early fetal rat brain upon acute and chronic ethanol exposure: Down regulation of PKA and p‐CREB expression.png)
Synapse
SynapseIF0
PREAI
ShareSaveShareSaveShareSaveShareSave
View more