Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/193122
Title: | Integrated Proactive and Reactive Failure Handling Strategies to Improve the Services in Computational Grid |
Researcher: | Latchoumy, P |
Guide(s): | Khader, P Sheik Abdul |
Keywords: | Integrated Proactive, Reactive, Failure Handling, Computational Grid |
University: | B S Abdur Rahman University |
Completed Date: | 2018 |
Abstract: | This study focuses on a computational grid which enables global sharing of the extensive computing power of heterogeneous resources for problem-solving in a large class of engineering, scientific and demand-driven commercial domains. This study develops grid resource management with failure handling services and it is one of the most challenging issues in a highly dynamic, heterogeneous and failure-prone grid environment. The primary function of Grid Resource Management System (GRMS) is to provide suitable resources for handling failures before and after scheduling the user job to complete successfully within the deadline. The present grid environment helps grid users to solve the computational and data-intensive applications without giving a guaranteed service. This work gives an assurance to the Quality of Service (QoS) requirements such as Deadline, Cost, Reliability, and Availability. By using Service Level Agreement (SLA), QoS can be made use of grid computing attractive for commercial applications. newlineThis study tries to develop an Integrated Proactive and Reactive Failure Handling Framework to deliver the guaranteed service to the grid users which may minimize the runtime failures and job migrations. It allocates the best suitable resources to run a job and manage the failures like resource, process, match-making, and time-out. It returns the job successfully to the grid user within the stipulated budget and deadline in the end. This framework discusses group scheduling strategy for failure management which enables to minimize the communication time and finally it proposes adaptive failure prevention methods to minimize the number of reschedulings and job migrations. This work further evaluates different components of the integrated framework using different strategies and evaluated using the performance parameters such as resource utilization, job success rate and job completion time. This framework is also experimented using grid simulator GridSim toolkit 5.2 and verified with the existing Min-M |
Pagination: | 133 |
URI: | http://hdl.handle.net/10603/193122 |
Appears in Departments: | Department of Computer Science and Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
abstract.pdf | Attached File | 19.07 kB | Adobe PDF | View/Open |
bibolography.pdf | 31.07 kB | Adobe PDF | View/Open | |
chapter 1.pdf | 184.72 kB | Adobe PDF | View/Open | |
chapter 2.pdf | 210.29 kB | Adobe PDF | View/Open | |
chapter 3.pdf | 685.63 kB | Adobe PDF | View/Open | |
chapter 4.pdf | 761.55 kB | Adobe PDF | View/Open | |
chapter 5.pdf | 671.68 kB | Adobe PDF | View/Open | |
chapter 6.pdf | 528.05 kB | Adobe PDF | View/Open | |
chapter 7.pdf | 550.01 kB | Adobe PDF | View/Open | |
chapter 8.pdf | 25.76 kB | Adobe PDF | View/Open | |
references.pdf | 210.7 kB | Adobe PDF | View/Open | |
table of contents.pdf | 80.9 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: