Loading...

Cost Efficient Task Scheduling Algorithms in Grid Environment

Kardani Moghaddam, Sara | 2012

494 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 42668 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Movaghar Rahimabadi, Ali
  7. Abstract:
  8. Computational grids enabling resource sharing and coordination are now one of the common and acceptable technologies used for solving computational intensive applications rising in scientific and industrial problems. Nevertheless, due to the heterogeneity, dynamicity and autonomy of the grid resources, task scheduling within these systems has become a challenging research area. Applying the market model to the grids is a good approach which can easily take the dynamic characteristics of the grid resources into account and simplify the scheduling problem considering user-centric factors.In this thesis a two level market-based model for task scheduling problem in grid environment is presented. The proposed model is composed of multiple sites; each site has multiple resources with different capabilities. A grid scheduler is responsible for assigning tasks to the best sites with the aim of executing tasks in most economical manner. A linear programming model is constructed and the best site for each task is selected. Each of these sites can have different access policies for their resources. In order to assign tasks to the best resources in each site, a hybrid Genetic algorithm and Variable Neighbourhood Search (VNS), named GA-VNS, is presented. The main objective of the proposed algorithm is to reduce the overall cost of task executions without any significant increment in system makespan. Performance of the proposed algorithm is evaluated against other algorithms. Simulation results obtained from performance evaluation show that GA-VNS performs much better than other algorithms in terms of cost of task executions. In addition, the proposed model is implemented using Matlab and GAMS software. The performance of the model is evaluated in different cases considering heterogeneity in tasks and resources
  9. Keywords:
  10. Scheduling ; Computational Grids ; Economy Model ; Genetic Algorithm ; Local Search

 Digital Object List

 Bookmark

No TOC