الفهرس 
Grid Technology and Review of Earlier Work
Disadvantages of grid computingOnly 14 pages are availabe for public view
 |  | from | 201 |  |  |
| | | from | 201 | | |
Abstract
The aim of the grid computing is to collect the unused resources into a single system that will be able to solve problems that are too complex for a single machine. Scheduling which plays an important role to measure the effectiveness of resources management to achieve high performance on grid computing environment is investigated. Developing scheduling algorithms for grid computing that involves some challenges due to the heterogeneity and highly dynamic nature of grid is introduced. One of the important challenges in the scheduling is to predict dynamically the run time of an application. A proposed dynamic performance prediction model which combines both the resource-oriented prediction and the application-oriented prediction characteristics is obtained based on running time prediction algorithm. A dynamic state of the resources which predicts the task run time is used in this work to obtain a time optimization scheduling algorithm based on back-propagation neural networks. The main objectives of this work are to reach both the minimization of tasks total completion time and the maximization of the throughput based on the preceding proposed model. The proposed task run time prediction model which introduces reliable information to the proposed scheduling algorithm is presented. This model predicts the submitted task’s run time by training neural networks through a training set of samples.