No abstract

No abstract

Reliable processing capacity and flexible storage space make Cloud computing the most recent favourable technology. Many organizations have converted their conventional processing data centre to a Cloud data centre. Cloud computing provides promising execution and storage, which leads to massive growth in processing demand by Cloud users. It makes the Cloud data centre increase the number of virtual machines (VM) to execute the users tasks. Hence, it causes high frequency disbursed and has increased energy consumption. Many techniques were proposed, which focuses on Cloud energy saving. However, there is still a lack of trade‐off between energy‐efficient task allocation and frequency scaling for a given workload. In this work, we propose a task scheduling algorithm that aims to minimize energy consumption through the frequency scaling technique while improving task execution time. Specifically, our scheduler comprises two modules, which are the scaling frequency module and frequency‐aware task scheduling module. In our first module, we utilize Dynamic Voltage and Frequency Scaling‐Optimal Frequency (DVFS) to determine the optimal frequency and selecting the best server for the incoming tasks. The number of VM is created upon the best server. As for the second module, the VM processing capacity is scaled to the required frequency of the task. We identify it as a required processing capacity for executing the tasks. The experiment result shows that our algorithm has outperformed and efficiently minimized the energy consumption in the Cloud data centre as compared with existing energy‐saving techniques. Meanwhile, the task allocation also has met the system's Quality of Service (QoS). Significantly, leveraging the resource processing frequency is able to gain better trade‐off between performance and energy consumption in the Cloud data centre.

Interest in broadband wireless access (BWA)

No abstract