Abstract
Objective: To study and analyze the existing scheduling algorithms i.e. by using Metaheuristic scheduling algorithms for effective resource provisioning in cloud computing. Methods: Metaheuristic scheduling algorithms for scheduling in cloud computing as they have faster convergence rather than the other optimization algorithms. The existing algorithms considered only few metrics like make span, resource utilization, processing cost, execution cost. Findings: In the existing scheduling algorithms, even though they have used the above-mentioned metrics still the resources in the cloud were not efficiently provisioned as the workload in the cloud is heterogeneous and dynamic. We have identified the few metrics which were not addressed by the previous authors i.e. electricity price per unit, average migration time and memory utilization, communication cost. Improvements: In the existing algorithms, few parameters were considered and still some of the parameters need to be addressed so as to schedule the resources in an effective way. In this paper, we have studied and analyzed about some of the task scheduling and workflow scheduling algorithms in cloud computing and given few research directions. Keywords: Cloud Computing, Makespan, Metaheuristic, Resource Utilization, Task Scheduling
Highlights
According to the definition of NIST, cloud computing can be defined as a paradigm for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources
Scheduling can be effected by the following metrics in cloud computing
We have studied about the different Metaheuristic scheduling algorithms in cloud computing as they are having faster convergence towards the global solution as the scheduling in cloud computing is a NP hard problem
Summary
According to the definition of NIST, cloud computing can be defined as a paradigm for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources. The major services of the cloud can be divided into three categories In this service, cloud users can be given the access to the services like compute, storage and Network on demand by using virtualized infrastructure. The main player of the IaaS services in the market is Amazon AWS By using this service users can get rid of the upfront costs on the physical infrastructure. In this service, cloud users can be able to access the PaaS service to develop new applications with the help of the platform provided by the cloud. Scheduling can be effected by the following metrics in cloud computing.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have