Design and Development of Dual Fuzzy Technique to Optimize Job Scheduling and Execution Time in Cloud Environment

Title

Design and Development of Dual Fuzzy Technique to Optimize Job Scheduling and Execution Time in Cloud Environment

Creator

Thingom, Chintureena

Contributor

R, Ganesh Kumar.

Description

Cloud computing is a type of computing that relies on sharing a pool of computing resources, rather than deploying local or personal hardware and software. It enables convenient, on-demand network access to a shared pool of configurable computing re- sources (e.g., applications, storage, networks, services, and servers) that can be swiftly provisioned and released with minimal management control or through the interaction of the cloud service provider. The increasing demand for computing resources in the cloud has made elasticity an important issue in the cloud. The availability of extending the resources pool for the user provides an effective alternative to deploying applications with high scalability and processing requirements. Providing a satisfactory Quality of Service (QoS) is an important objective in cloud data centers. The QoS is measured in terms of response time, job completion time and reliability. If the user jobs cannot be executed in high load and the job is crashed, it will enormously increase the response time and also push up the job completion time. Also due to load, the jobs may be still in the waiting queue and could not find a resource to execute. In such a situation, the user notices a big response delay and it will affect the QoS. Towards ensuring QoS, this research proposes the following solution - Dual Fuzzy Load Balancing for jobs. Dual Fuzzy Load Balancing balances the load in the data center with an overall goal of reduction of response and execution time for tasks. The proposed solutions were simulated in the Cloudsim simulator and performance metrics in terms of job response time, job completion time, resource utilization, a number of SLA violations, and along with the cost comparison to the existing algorithms of Load Balancing. The proposed solutions are also implemented in a real cloud environment and the effectiveness of the solution is evaluated.

Source

Author's Submission

Date

2022-01-01

Publisher

Christ(Deemed to be University)

Subject

Computer Science and Engineering

Rights

Open Access

Relation

61000209

Format

PDF

Language

English

Type

PhD

Identifier

http://hdl.handle.net/10603/453657