Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.715889
Title: Workflow framework for cloud-based distributed simulation
Author: Chaudhry, Nauman Riaz
Awarding Body: Brunel University London
Current Institution: Brunel University
Date of Award: 2016
Availability of Full Text:
Access through EThOS:
Full text unavailable from EThOS. Restricted access.
Access through Institution:
Abstract:
Although distributed simulation (DS) using parallel computing has received considerable research and development in a number of compute-intensive fields, it has still to be significantly adopted by the wider simulation community. According to scientific literature, major reasons for low adoption of cloud-based services for DS execution are the perceived complexities of understanding and managing the underlying architecture and software for deploying DS models, as well as the remaining challenges in performance and interoperability of cloud-based DS. The focus of this study, therefore, has been to design and test the feasibility of a well-integrated, generic, workflow structured framework that is universal in character and transparent in implementation. The choice of a workflow framework for implementing cloud-based DS was influenced by the ability of scientific workflow management systems to define, execute, and actively manage computing workflows. As a result of this study, a hybrid workflow framework, combined with four cloud-based implementation services, has been used to develop an integrated potential standard for workflow implementation of cloud-based DS, which has been named the WORLDS framework (Workflow Framework for Cloud-based Distributed Simulation). The main contribution of this research study is the WORLDS framework itself, which identifies five services (including a Parametric Study Service) that can potentially be provided through the use of workflow technologies to deliver effective cloud-based distributed simulation that is transparently provisioned for the user. This takes DS a significant step closer to its provision as a viable cloud-based service (DSaaS). In addition, the study introduces a simple workflow solution to applying parametric studies to distributed simulations. Further research to confirm the generic nature of the workflow framework, to apply and test modified HLA standards, and to introduce a simulation analytics function by modifying the workflow is anticipated.
Supervisor: Taylor, S. ; Bell, D. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.715889  DOI: Not available
Share: