Host Utilization Algorithm to Reduce Energy Consumption in Cloud Computing
S. Kanagasubaraja1, Hema M2, Vineeth M3, Shreyas Rao K4, Siva Anandh S5

1Dr. S. Kanagasubaraja*, Easwari Engineering College, Chennai, India.
2Hema M, Easwari Engineering College, Chennai, India.
3Vineeth M, Easwari Engineering College, Chennai, India.
4Shreyas Rao K, Easwari Engineering College, Chennai, India.
5Siva Anandh S, Easwari Engineering College, Chennai, India.

Manuscript received on April 11, 2020. | Revised Manuscript received on May 15, 2020. | Manuscript published on June 30, 2020. | PP: 352-359 | Volume-9 Issue-5, June 2020. | Retrieval Number: E9547069520/2020©BEIESP | DOI: 10.35940/ijeat.E9547.069520
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: The utilization of distributed computing server farm is developing quickly to fulfill the large increment required for systems administration, High-Performance Computing(HPC) as well as stockpiling assets for executing business and logical applications. The process of Virtual Machine (VM) solidification is inclusive of VMs getting relocated in order to make use of less physical servers. As a result, it enables the shut down or low-power mode of more number of servers which enhances the vitality utilization effectiveness, working expense and CO2 discharge. An urgent advance in VM union is have over-burden discovery, which endeavors to foresee whether a physical server is going to be oversubscribed with VMs. On the contrary to usual studies which performed utilization of CPU being the standalone indicator for host overload, a multiple correlation host overload detection algorithm was proposed in the recent study by considering a lot of factors in this regard. A higher load balance model was introduced in this text for the general public cloud, supported by the concept of cloud partitioning, in addition to a switch mechanism used to strategize differently under different scenarios. The IP address is generally shared by a true server and carbo balance. In this regard, the load balancer considers the interface developed with IP address which accepts request packets and the packets are directed to the selected servers. With an aim to improve the efficiency in public cloud environment, the algorithm employed the sport theory in the load balancing strategy. 
Keywords: CO2, CPU, Virtual Machine.t