Improving Task Allotment by using Deadline Resource Provisioning and Semo Algorithm in Cloud Computing
S. Chitra
S. Kamalee
S. Sasidharan
M. Sohail Ahmed
K. Balamurugan
Keywords: Task Allotment, Cloud Computing, Provisioning Mechanism, Resource.
Abstract
One of the most recent developments in computing is cloud computing, which makes it simple for anyone to access the Internet. Numerous Cloud services rely on users of the Cloud to virtualize Cloud software and map it. Typically, the requests made by Cloud users from multiple terminals result in busy data centres and servers and imbalanced workloads. Therefore, all cloud servers must have a Cloud job allocation that employs an effective task allocation technique. This thesis suggests an algorithm called the deadline resource provisioning algorithm that optimises job allocation across the many virtual machines in the Cloud server, boosting the response efficiency of Cloud servers while also increasing accuracy. Methods and discoveries for resolving task imbalance concerns in cloud servers. Furthermore, our method will show that using an effective task allocator technique improves throughput and response time in mobile and cloud contexts.
This project presents a technique for resource provisioning and scheduling for scientific workflows on Infrastructure as a Service (IaaS) and Platform as a Service clouds (PaaS). This study provides a Superior Element Multitude Optimisation (SEMO)-based technique that attempts to reduce overall workflow execution costs while fulfilling deadline limitations. The primary goal of the project is to analyse the best available resource in the cloud environment based on total execution time and total execution cost when comparing one process to another. Dominant firefly behaviour is applied to Cloud job allotment strategies in the present system, which is referred to as the dominant firefly algorithm. There will be numerous dominant fireflies and many submissive fireflies in a group of fireflies, and it has downsides such as being adaptable only in scenarios with the same initial set of resource availability. Only works in a single cloud service provider environment. The cost of data transport between cloud data centres is not considered.
In addition to the present system implementation, the Deadline resource provisioning mechanism for Job execution is being implemented. The dissertation presented the SEMO (Superior Element Multitude Optimization) algorithm, which compares the overall execution time and total execution cost of one operation to another. Adaptable in situations where a diverse initial set of resources is available. Appropriate for a variety of cloud service provider scenarios. The cost of data transfer between cloud areas is lowered. The performance metrics of the allotment mechanism quantify its efficiency in terms of time complexity and space complexity.
