Authors: Kun Huang; Ning Han
Addresses: School of Computer and Communication, Hunan Institute of Engineering, Xiangtan City, 411100, China ' Department of Networking Engineering, Hunan Institute of Engineering, Xiangtan 411104, China
Abstract: As more and more scientific workflows have been deployed and executed in cloud environments, how to optimise the execution efficiency and quality of workflow applications has become an important issue that needs to be addressed. Although many workflow management systems have been proposed in recent years, the performance bottleneck when running scientific workflows is still the low-efficiency of data-related services. In this paper, we present a workflow-oriented optimisation plug-in system, namely data-service enhanced middleware for scientific workflow (DSEM-SW), which provides a set of mechanisms to improve the efficiency and quality of data processing when running scientific workflows. Extensive experiments are conducted to evaluate the performance of the proposed middleware in a real-world cloud test-bed. The experimental results indicate that our DSEM-SW outperforms many existing systems in terms of different performance metrics, and can significantly improve the execution efficiency of data-intensive workflows.
Keywords: workflow; cloud computing; resource virtualisation; quality of data; QoD.
International Journal of Networking and Virtual Organisations, 2021 Vol.24 No.2, pp.126 - 143
Received: 19 Dec 2019
Accepted: 02 Mar 2020
Published online: 21 Apr 2021 *