Cost Optimised Heuristic Algorithm (COHA) for Scientific Workflow Scheduling in IaaS Cloud Environment

Cloud computing, a multipurpose and high-performance internet-based computing, can model and transform a large range of application requirements into a set of workflow tasks. It allows users to represent their computational needs conveniently for data retrieval, reformatting, and analysis. However, workflow applications are big data applications and often take long hours to finish executing due to their nature and data size. In this paper, we study the cost optimised scheduling algorithms in cloud and proposed a novel task splitting algorithm named Cost optimised Heuristic Algorithm (COHA) for the cloud scheduler to optimise the execution cost. In this algorithm, the large tasks are split into sub-tasks to reduce their execution time. The design purpose is to enable all tasks to adequately meet their deadlines. We have carefully tested the performance of the COHA with a list of workflow inputs. The simulation results have convincingly demonstrated that COHA can effectively perform VM allocation and deployment, and well handle randomly arrived tasks. It can efficiently reduce execution costs while also allowing all tasks to properly finish before their deadlines. Overall, the improvements in our algorithm have remarkably reduced the execution cost by 32.5% for Sipht, 3.9% for Montage, and 1.2% for CyberShake workflows when compared to the state of art work.

I. Introduction

Cloud computing is a multi-tenant internet-based platform that provides digital service delivery by various cloud service providers on-demand base which is subject to quality of service (QoS) constraints [1], [2], [3]. We have seen a rapid growth in cloud technology development in recent years. This is due to its vast benefits including 1) cost reduction in upfront capital expenditure on hardware, software, hosting and deployment; 2) availability; 3) location independence; 4) flexibility and 5) market agility [4]. Cloud provides: (i) softwares which are available via a third-party over the Internet referred as software as a service (SaaS), (ii) services such as storage, networking, and virtualization known as infrastructure as a service (IaaS) and (iii) hardware and software tools available over the internet which is commonly referred to as platform as a Service (PaaS) [5].