Date of Award

8-1-2019

Thesis Type

masters

Document Type

Thesis (Restricted Access)

Divisions

fsktm

Department

Faculty of Computer Science & Information Technology

Institution

University of Malaya

Abstract

Virtualization allows multiple operating systems and applications to be executed on the same physical server concurrently. Recently, two popular virtualization platforms, namely container-based and hypervisor-based were adapted into most data centers to support cloud services. With the increase in various types of scientific workflow applications in the cloud, low-overhead virtualization techniques are becoming indispensable. However, to deploy the workflow tasks to a suitable virtualization platform in the cloud is a challenge. It requires intimate knowledge of ever-changing workflow tasks at any given moment. This research proposed an automated system architecture that can choose the best virtualization platform to execute workflow tasks. A benchmark performance evaluation was conducted on various workflow tasks running on container-based and hypervisor-based virtualization. Several tools were used to measure the metric, such as central processing unit (CPU), memory, input and output (I/O). Based on the benchmark, a system architecture was created to automate the virtualization platform selection. The results showed that the proposed architecture minimized the workflows’ total execution time.

Note

Dissertation (M.A.) – Faculty of Computer Science & Information Technology, University of Malaya, 2019.

Download

Share

COinS