An efficient green computing fair resource allocation in cloud computing using modified deep reinforcement learning algo
- PDF / 2,179,617 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 24 Downloads / 220 Views
(0123456789().,-volV)(0123456789(). ,- volV)
METHODOLOGIES AND APPLICATION
An efficient green computing fair resource allocation in cloud computing using modified deep reinforcement learning algorithm K. Karthiban1 • Jennifer S. Raj2
Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Cloud computing provides services and resources in the Internet, and many applications are self-service-supported, ondemand resource allocation-adapted. These dynamic networks allocated necessary resource to the users’ need and they require proper resource allocation scheme. Since various resources are consumed by users if resource allocation is not proper, this leads the system to load imbalance nature. Using Internet-connected devices for storage and computation not only communicates the cloud resources but also connects the devices to network through various protocols. These changes make the network into a complex, dense, heterogeneous system. In this paper, a green computing fair resource allocation through deep reinforcement learning model is proposed to provide efficient resource allocation scheme to the users in the network. Conventional Q-learning model fails in dimensionality problem when the state space increases exponentially. The proposed model is combined with fair resource allocation with deep reinforcement learning to provide better allocation schemes compared to the conventional model. Keywords Deep reinforcement Fair resource allocation Cloud computing
1 Introduction Cloud computing is an emerging technology which is used in many applications including data analysis, storage and Internet of things (IoT). In the field of distributed computing, the traditional methods use enterprises to handle services and cloud computing changes by providing different types of services to users. All these services are web services, and the user does not need to invest in computing environment. Cloud computing provides services such as • Infrastructure as a service (IaaS) • Platform as a service (PaaS) • Software as a service (SaaS) Communicated by V. Loia. & K. Karthiban [email protected] Jennifer S. Raj [email protected] 1
Department of Information Technology, Karpagam College of Engineering, Coimbatore, Tamilnadu, India
2
Department of ECE, Gnanamani College of Technology, Namakkal, Tamilnadu, India
In the above services, the users need to submit request to obtain the service from the service provider through Internet and the service provider is responsible for allocating and maintaining the resources to fulfill the user requests. Service providers use scheduling and resource allocation algorithms which schedule the incoming requests and manage the resource, so that allocating resource by managing them in full utilization maximizes the revenue. The major terms in cloud computing in realtime environment are resource allocation and scheduling. Figure 1 gives an illustration about the general cloud computing model. The allocation of resource to users is a crucial objective in cloud computing sinc
Data Loading...
