Toggle Main Menu Toggle Search

Open Access padlockePrints

Virtual Machine Live Migration in Trace-driven Energy-Aware Simulation of High-Throughput Computing Systems

Lookup NU author(s): Osama Alrajeh, Dr Matthew ForshawORCiD, Dr Nigel Thomas

Downloads


Licence

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND).


Abstract

Virtual Machine (VM) live migration is one of the strategic approaches that can be employed to reduceenergy consumption and increase the utilisation of a large computing infrastructure. In this paper, weimplement the pre-copy live migration algorithm to provide a test environment for live job migrationin a high throughput computing system (HTC). We present details of our extension into the HTCSim simulation framework. Furthermore, we demonstrate the performance and energy impact ofour responsive live migration approach in high throughput computing (HTC) systems with variousmigration policies. The proposed policies focus on which computer to select for migration from anHTC resource pool. We evaluate our migration policies through the use of our HTC-Sim simulationframework. Moreover, we compare the results between the policies as well as the system wheremigration is not considered. We demonstrate that our responsive migration could save approximately75% of the system wasted energy due to job evictions by user interruptions where migration is notemployed as a fault-tolerance mechanism.


Publication metadata

Author(s): Alrajeh O, Forshaw M, Thomas N

Publication type: Article

Publication status: Published

Journal: Sustainable Computing: Informatics and Systems

Year: 2020

Volume: 29

Issue: B

Print publication date: 01/03/2021

Online publication date: 04/11/2020

Acceptance date: 18/10/2020

Date deposited: 03/11/2020

ISSN (print): 2210-5379

ISSN (electronic): 2210-5379

Publisher: Elsevier

URL: https://doi.org/10.1016/j.suscom.2020.100468

DOI: 10.1016/j.suscom.2020.100468


Altmetrics

Altmetrics provided by Altmetric


Share