Open AccessEnhanced Resource Allocation and Workload Management using Reinforcement Learning Method for Cloud Environment
Author(s) -
P. Suresh,
P. Keerthika,
K. Logeswaran,
R. Manjula Devi,
M. Sangeetha
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.d8983.118419
Subject(s) - cloud computing , computer science , workload , server , scalability , distributed computing , reinforcement learning , scheduling (production processes) , virtual machine , resource allocation , live migration , resource management (computing) , computer network , virtualization , database , operating system , artificial intelligence , operations management , engineering
Cloud computing is a delivery model of IT resources such as computing servers, storage, databases, networking and software over the Internet. It offers the resources as services based on demand with more faster, flexible and economies of scale. The major challenges in the cloud computing are resource allocation and workload management due to the scalability of the cloud users and the services deployed in it. Even though there are various approaches available to manage workload and resource allocation, unfortunately most of them fail to mange it properly. This paper proposes a Reinforcement Learning based Enhanced Resource Allocation and Workload Management (RL-ERAWM) approach to increase the performance of cloud with large number of tasks and users. It implements the Q-Learning approach which effectively considers arrival rate of the requests and workload of the virtual machine. Experimental results prove that the proposed method alleviates the performance of task scheduling and workload management process compared with other approaches in terms of response time, makespan and virtual machine utilization.