A parallel solution for scheduling of real time applications on grid environments

  • Authors:

  • Mohammad Kalantari;Mohammad Kazem Akbari

  • Affiliations:

  • Computer Engineering Department, Amirkabir University of Technology (Tehran Polytechnic), Tehran, Iran;Computer Engineering Department, Amirkabir University of Technology (Tehran Polytechnic), Tehran, Iran

  • Venue:
  • Future Generation Computer Systems
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

One of the objectives of computational grids is to offer applications with the collective computational power of distributed but typically shared heterogeneous resources. Unfortunately, efficiently harnessing the performance potential of such systems (i.e., scheduling) is a challenging endeavor due to the principally distributed, shared and heterogeneous nature of the resources involved. The main focus of this work is devoted to online scheduling of real time applications in distributed environments such as grids. Specifically, we are interested in applications, with several independent tasks, and each task labeled with a lifecycle called deadline. Here, our goal is to schedule applications within specified deadlines. To achieve this, performance dynamism of resources is modeled with the help of queuing techniques. A mathematical neural model is employed afterwards, to schedule subtasks of the application. The main contribution of this work is an efficient and fast parallel scheduling algorithm under time constraint and heterogeneous resources, appropriate for implementation in parallel machines in O(1) time. The proposed model was implemented on GridSim toolkit and under various configurations and parameters to evaluate the performance of scheduling algorithms. Simulation experiments have shown that in approximately 96% of cases, our model schedules tasks successfully and efficiently.