US-Egypt Cooperative Research: Real-Time Scheduling of Task Graphs on Grid Computing

University Of Connecticut, Storrs CT

Investigators

Abstract

0203293 Ammar Description: This award is to support a collaborative project by Dr. Reda A. Ammar, Department of Computer Science and Engineering, University of Connecticut, Storrs, Connecticut and Dr. Ayman El Dessouki, President of the Electronics Research Institute, Cairo, Egypt. They plan to study real-time scheduling of task graphs on grid computing. The investigators will focus on developing a set of new techniques and heuristics to schedule real-time applications on a grid. They will assume that each application consists of a set of connected tasks. The objective of this work is to meet the requirements of task deadlines, to improve the processing power utilization, and to increase the throughput. Their scheduling approaches utilize the available processing power on each processor to accommodate as many tasks of different applications as possible while satisfying the required deadline of each task. The algorithms reduce the communication cost among tasks and the possibility of processing power fragmentation. They will demonstrate that the proposed scheduling technique produces an adequate acceptance rate compared to other approaches. Finally, evaluation of the research results will be through initial simulation studies, followed by experimental testing on an available grid. Scope: Scheduling a large number of high performance computing applications on a grid computing environment is a serious obstacle to achieving a good performance, and this becomes more critical in real time systems. A grid scheduler without enough knowledge of the state of the Grid and the scheduled tasks of a given application cannot adequately manage the Grid resources, may fragment the available processing power and introduce a high volume of communication among the application's tasks, and may cause rejection of some submitted applications if some of its tasks miss their deadlines. This project seeks to provide mechanisms to satisfy the performance requirements while maximizing the processing power utilization, and thus advance the field of grid computing. Applications that are computationally extensive, such as weather prediction, numerical simulations in physics, biology and chemistry, geodynamics, telecommunications, real-time image recognition, e-business, e-engineering will benefit from the proposed scheduling algorithms. The two PIs had prior collaboration and their institutes are well equipped for the proposed project. The project will involve U.S. and Egyptian graduate students. This project is being supported under the US-Egypt Joint Fund Program, which provides grants to scientists and engineers in both countries to carry out these cooperative activities.