Cloudcomputing Overview
Now a day the group of the science community and industrial community peoples are using cloud computing services. The cloud computing services are offered their infrastructure at low cost.The companies no need to purchase or sustain their personal computing infrastructure, that is the advantage of the cloud computing (Beloglazov & Buyya, 2013).
Cloud Computing is one of the best research directions, it’s the aggregate of utility computing, grid computing and distributed computing. The data processing and storages are highly increasing in the current computing world. So it guides the expressive growth of Data forms (Data Centers). Cloud Data centers offer various the types of services. It provides resource as a service that is an Information Technology (IT) Server, Networking components in the data center, Storage, Cooling system, etc., those hardware components are coming under the Infrastructure as a Service (IaaS). Software Components are such as social networks, search engine, software tools, Computing applications, etcare consider, as Software as a Service (SaaS) (Dong et al., 2015).
Objective
Cloud computing technology has resulted in maintaining large-scale Data centers consisted of thousands of computing nodes that consume ample amounts of electricity. In future energy consumption will be increased 53% compared to today’s consumption.
In another report said that, data center server’s power consumption is only between 10% to 15% of supply electricity.One of the primary reasons for the high electricity consumption is a incapable usage of the Data center resources. Due to the slender dynamic power vary of servers, it's seen that, even idle servers consume regarding seventy you look after their peak power (Fan et al., 2007). Idle servers consume more than 70% of energy consumption compared to active server. The proper server load balancing improves the resource utilization. Efficient resource usage increases the idle server count. And then ideal servers put into as switch off mode. The power off method of idle server increases the energy efficiency among the servers in the cloud data center. Other related factors to the energy consumption are delay and Quality of Service (QoS).
In order to scale down the energy consumption in cloud data center, this paper proposes three main tasks, the primary one is characteristic the load balancing factors and redistribute the load, the other is find out the most suitable server, to assigning the task to the server, achieved by Most Efficient First Fit Algorithm (MEFFA)(Innocent et al., 2018) and also the third algorithm is processed the task within the server in an efficient way by Energy efficient Virtual Round Robin (EEVRR) Scheduling Algorithm. To boot this paper, focuses FAT Tree topology architecture in Data Centers.
The best task scheduling approach lower the energy consumption, improves the resource usage, shorten the make span time and it additionally cuts down the delay among cloud servers. Scheduler has logic to seek out the most suitable server and assign the task to the server, based on the EEVRR algorithm. Another one task is load balancing that load balancer balance the load on the servers within the data centers. Task transferred from extremely used server to the lowly utilized server and a few of the underutilized server’s that means that the tasks redirected to the excellent servers. Currently the idle servers are turned off. This paper compares the proposed algorithm with Round Robin (RR) and Weighted Round Robin (WRR) Algorithm. Experimental results show the proposed algorithm performance. The remaining part of the paper is, section 2 mentions the related work, in section 3 discusses proposed work, section 4 shows the mathematical model of the paper, section 5 shows the Experimental results and comparison, and section 6 discussed Conclusion and section 7 addresses the reference.
In(Yavari et al., 2019) “Maede Yavari” et al followed the heuristic and Meta –heuristic algorithm were followed to improve the energy efficiency. The heuristic was named as HET-VC(Heuristic energy and VM consolidation based on temperature) and the meta heuristic was FET-VC (FireFly Energy and VM Consolidation based on temperature). The experiments had conducted with the Cloud Sim simulator. In the final conclusion the author achieved the 76% energy efficiency through SLA.