Abstract
Real-Time Embedded Systems (RTES) must be verified for their timing correctness where knowledge about the
Worst-Case Execution Time (WCET) is the building block of such verification. Traditionally, research on the
WCET analysis of RTES assumes sequential code running on single-core platforms. However, as computation is
steadily moving towards using a combination of parallel programming and multi-core hardware, new challenges
in timing analysis, and especially in WCET analysis need to be addressed. Towards this direction, this paper
presents the Timing Analysis tool for Parallel Embedded Software (TAPES). The proposed tool allows the WCET
estimation of parallel applications running on multi-core hardware through a hybrid measurement-based
analysis method, that combines the program flow and timing information into an Integer-Linear Programming
problem to estimate the WCET. In addition, the TAPES tool allows the measurement of the longest end-to-end
execution time by capturing the timing properties of the parallel executing threads using time-stamped execution
traces of the program. The applicability of the proposed tool is demonstrated through the timing analysis of an
embedded parallel benchmark suite – the ParMiBench. The results showed that the calculated WCET estimates
have significantly less over-approximation compared to the measured WCET estimates. The comparison of the
calculated and measured WCET estimates showed modest over-estimates.
Muhammad Waqar Aziz, Syed Abdul Baqi Shah. (2018) Timing Analysis of Parallel Software Applications for Multi-core Embedded Systems, Pakistan Journal of Engineering and Applied Sciences, VOLUME 22, Issue 1.
-
Views
2006 -
Downloads
192
Article Details
Volume
Issue
Type
Language
