Browse by author
Lookup NU author(s): Sami Alajrami,
Professor Alexander Romanovsky
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
Copyright © 2018 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved. Using cloud computing to execute software processes brings several benefits to software development. In a previous work, we proposed a reference architecture, which treats software processes as workflows and uses cloud computing to execute them. Scheduling the execution in the cloud impacts the execution cost and the cloud resources utilization. Existing workflow scheduling algorithms target business and scientific (data-driven) workflows, but not software processes workflows. In this paper, we adapt three scheduling algorithms for our architecture and propose a fourth one; the Proportional Adaptive Task Schedule algorithm. We evaluate the algorithms in terms of their execution cost, makespan and cloud resource utilization. Our results show that our proposed algorithm saves between 19.74% and 45.78% of the execution cost and provides the best resource (virtual machine) utilization compared to the adapted algorithms while providing the second best makespan.
Author(s): Alajrami S, Romanovsky A, Gallina B
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Published
Conference Name: MODELSWARD 2018 - Proceedings of the 6th International Conference on Model-Driven Engineering and Software Development
Year of Conference: 2018
Online publication date: 22/01/2018
Acceptance date: 22/01/2018
Library holdings: Search Newcastle University Library for this item