Efficient Processing Power Reservation Approach to Imporove Real-Time Task Schedulability and Relaibaility
Efficient utilization of the computational resources is an urgent demand especially if real time constraints should be guaranteed. Moreover, an acceptable level of reliability should be provided due to the critical nature of some real time applications. This paper proposes a new approach for processing power reservations that efficiently utilizes all the available processing power to improve reliability and schedulability of independent real time tasks on a uni-processor. The basic idea of the proposed approach is to use all of the available processing power in the time interval between the arrivals of two successive tasks or when an existing task departs. The advantages of this mechanism are: 1) it reduces the execution time required for each task and hence increases its reliability. 2) At the arrival of a new task; the processing power requirements for the executing tasks to meet their deadlines are smaller, which gives the new arriving tasks higher chances to be accommodated with the existing ones. 3) Efficient processing power utilization may reduce the power consumption in processors with dynamic voltage scaling. An illustration example and simulation experiments showed that our approach provides a more reliable scheduling scheme with higher acceptance rate compared to the traditional approach based on Rialto operating system.
Keywords: Processing Power, Processor Utilization, Scheduling, Real-Time Tasks
Download Full-Text
ABOUT THE AUTHORS
Ahmed E. Youssef
Faculty of Engineering-Helwan, Helwan University Cairo, Egypt, and Dept. of Information Systems, King Saud University Riyadh, 11543, KSA
Abeer Hamdy
Dept. of Computers and Systems, Electronic Research Institute, and Faculty of informatics and computer science, British University in Egypt Egypt
Reda Ammar
3Dep. of Computer Science and Engineering, University of Connecticut Storrs, CT 06269-3155, USA
Ahmed E. Youssef
Faculty of Engineering-Helwan, Helwan University Cairo, Egypt, and Dept. of Information Systems, King Saud University Riyadh, 11543, KSA
Abeer Hamdy
Dept. of Computers and Systems, Electronic Research Institute, and Faculty of informatics and computer science, British University in Egypt Egypt
Reda Ammar
3Dep. of Computer Science and Engineering, University of Connecticut Storrs, CT 06269-3155, USA