1.6 Previous study of Cloud Computing

      In cloud computing environment, a simulation-based approach to analyze CPU Debt in a Cloud is to be done for the minimization of waiting time in the queue and in the system for the purpose of use in the core research area in academia and in industry. It also deals with the number of servers or number of CPU to deliver the services as per the user request. If we get successful in minimizing the waiting time, then this will lead to customer satisfaction and usage of less number of CPU cores which will have influence on the performance of the cost optimization.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

Buyya, Ranjan, and Calheiros, 2009, have presented CloudSim toolkit which can enable to model and simulate in cloud computing architect. In CloudSim simulator, they have added the option for creating the number of Virtual Machines (VMs) in a particular Data Center, along with the help of different VM allocation and VM selection policies to model.

Pu, Liu, et al., 2010, have presented their performance analysis in parallel progression of CPU. They have also focused on monitoring workload on Xen Virtual Machine Monitors, which is concerned with network intensive workload. Paper focused on dealing with the experiments to find out the performance measurements on network I/O workload.

Khazaei, Misic, et al., 2011, have discussed about the techniques of resource provisioning, the rules of delivering different types of services such as infrastructure-based, platform-based, and software-based. Their research was concentrated to calculate the performance measurement while provisioning the resources to realize the Service Level Agreements (SLAs).

Spillner, Brito, et al., 2012, have presented an economically compensation concept to raise the detail and efficacy of reserved computation. This paper explains, how to enable highly virtualized resource broker in the business-oriented market place. This facilitates the consumer with configurable VMs for resource sharing. This paper supports on-demand resource provisioning with the help of scalability.

Khazaei, Misic, et al., 2012, have discussed about the modeling of cloud centers. They have proposed a performance measurement model to check the cloud farms and found out the solution to get the estimation of probability distribution.

Pal and Pattnaik, 2013, have presented virtualization classification in cloud computing environment. Virtualization technology manages and coordinates the accesses from the resource pool. Virtualization helps the CSPs to handle the composite workloads, and different software technology. They have also put the light on the virtualization classification and their working principle in the paper.

Xiao, Song, an Chen, 2013, have described the technique of allocating data center resources through virtualization technology, with this the idea of “skewness” for measuring the unevenness of resource use of a server in multidimensional way was originated. They have also developed a way by which the overall consumption of server resources can be improved.

Karthick, Ramaraj, and Subramanian, 2014, have proposed MQS (Multi Queue Scheduling) algorithm which aims to lower the cost of both on-demand requirements and reserved plans with the help of global scheduler, which intends to share the physical resources to its greatest level. The proposed algorithm uses the technique of clustering the tasks depending upon the burst time. This paper also reduces the chances of fragmentation and minimizes the starvation problem.

Yang, Kwon, et al., 2014, have introduced the techniques which requires compiler code analysis. This procedure minimizes the transferred data size with the help of changing the heap objects. They have discussed the procedure of cost cutting techniques for dynamic execution in cloud. Their result shows that reduced size affects both the transfer time and execution offloading in an efficient way.

Calero and Aguado, 2015, have presented a monitoring architecture concerned to the CSP and cloud user. This architecture allows the user to customize the metrics. The cloud providers can easily track the services used by the users.