Exact Schedulability Analysis of Global Multi-Processor IUF Scheduling using Symbolic Model Checking

Alok, Lele and Kumar, Ajith and Rohith, H M and Singh, Sanjay (2010) Exact Schedulability Analysis of Global Multi-Processor IUF Scheduling using Symbolic Model Checking. In: 5th International Conference on Industrial and Information Systems, ICIIS 2010, Jul 29-Aug 1, 2010, India.

[img] PDF
29071411.pdf - Published Version
Restricted to Registered users only

Download (1MB) | Request a copy
Official URL: doi: 10.1109/ICIINFS.2010.5578694


Real time scheduling spans a broad spectrum of algorithms for simple uniprocessors to highly sophisticated multiprocessors. As the world nears the end of Moore’s Law, more emphasis is being laid on the use of multi-processors in embedded systems. This has lead to a rise in several new scheduling schemes for multi-processor environments. Naturally there arises a need for an accurate comparison of the schedulabilities of each of these schemes in order to prove their superiority over others. The two-fold aim of this paper is to present a technique for the exact schedulability analysis of global multi-processor schedulers and, based on this intermediary result, prove the schedulability superiority of a novel multi-processor scheduling algorithm. The exact schedulability analysis technique presented here is symbolic model checking in NuSMV, and we have used this technique to investigate the schedulability of global Instantaneous Utilization Factor (IUF) based scheduling. Our analysis shows a substantial improvement in the acceptance of task sets with heavier utilizations.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Symbolic Model checking, Schedulability, Instantaneous Utilization Factor, scheduling, Multi-processor scheduling
Subjects: Engineering > MIT Manipal > Information and Communication Technology
Depositing User: MIT Library
Date Deposited: 31 Jul 2014 07:21
Last Modified: 31 Jul 2014 07:21
URI: http://eprints.manipal.edu/id/eprint/140287

Actions (login required)

View Item View Item