A Dynamic Threshold Based Energy Efficient Method for Cloud Datacenters

A Dynamic Threshold Based Energy Efficient Method for Cloud Datacenters

Shally (Banasthali Vidyapith, Rajasthan, India), Sanjay Kumar Sharma (Banasthali Vidyapith, Rajasthan, India) and Sunil Kumar (Manipal University Jaipur, Jaipur, India)
Copyright: © 2020 |Pages: 14
DOI: 10.4018/IJSI.2020040104
OnDemand PDF Download:
No Current Special Offers


Cloud computing has opened new avenues in the computing resource provisioning. It is providing affordable computing services to the users and opportunities to the cloud service providers to scale up their business. However, the upscaling resulted in the creation of huge datacenters running round the clock. Energy consumption in these datacenters is becoming a burning issue due to environmental hazards and operating costs. A novel method using dynamic threshold has been proposed in this article. Thresholds are used to migrate the virtual machines (VM) on physical machines (PM) for consolidation. A dynamic threshold selection method is used to reduce energy consumed by physical machines of the data centers. The upper and lower thresholds are set dynamically based upon the previous pattern of the CPU utilization. Experiments conducted on PlanetLab data show significant improvement in the energy efficiency without compromising SLA.
Article Preview


Cloud computing provides a large number of resources through the internet on pay as per use model (Buyya et al., 2009). Kleinrock (2005) has treated the cloud computing into utility computing. Actually, the idea of utility computing was given by John McCarthy in 1961 at MIT centennial celebrations:

If computer of the kind, I have advocated become the computers of the future, then computing may someday be organized as a public utility just as telephone system is a public utility the … The computer utility could become the basis of a new and important industry.

Recent advancements in software and hardware have made this dream come true in the form of cloud computing. There are various descriptions of cloud computing, although the standard definition was given by NIST (Mell & Grance, 2011):

Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.

Due to the cost-effective services provided by cloud computing, it has gained popularity these days. This resulted into an enlargement of the size of datacenters day by day. Huge data centers consist of a large number of physical resources and hence consumes very high amount of electricity. It has been found that the expense of energy consumption is nearly 12% of monthly operational expenditures of a typical data center (Gartner, 2010). With the current demand for cloud data center, power consumption is 0.5% of the world’s electricity usage and it will become 2% by 2020 (Forbes, 2015). So, the power consumption by the data centers is a burning issue not only due to electricity bills but it also poses a challenge to the environment in terms of carbon footprint. (Koomey, 2007) suggested that it’s necessary to manage cloud resources in energy efficient way to put a check on energy consumption and subsequent reduction in carbon footprint. Virtualization technology has played an important role in the reduction in energy consumption. Virtualization allows sharing of resources on a single physical machine. Multiple virtual machines can be created on single physical machine. With the help of it, multiple users can use a single physical machine as if their own machine without any intervention of others. Further improvement in the efficiency of energy consumption is achieved through virtual machine consolidation.

Complete Article List

Search this Journal:
Open Access Articles
Volume 10: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 9: 4 Issues (2021)
Volume 8: 4 Issues (2020)
Volume 7: 4 Issues (2019)
Volume 6: 4 Issues (2018)
Volume 5: 4 Issues (2017)
Volume 4: 4 Issues (2016)
Volume 3: 4 Issues (2015)
Volume 2: 4 Issues (2014)
Volume 1: 4 Issues (2013)
View Complete Journal Contents Listing