Change search
ReferencesLink to record
Permanent link

Direct link
Evaluation of a method for identifying timing models
Mälardalen University, School of Innovation, Design and Engineering. (Programming Languages)
2012 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

In today’s world, embedded systems which have very large and highly configurable software systems, consisting of hundreds of tasks with huge lines of code and mostly with real-time constraints, has replaced the traditional systems. Generally in real-time systems, the WCET of a program is a crucial component, which is the longest execution time of a specified task. WCET is determined by WCET analysis techniques and the values produced should be tight and safe to ensure the proper timing behavior of a real-time system. Static WCET is one of the techniques to compute the upper bounds of the execution time of programs, without actually executing the programs but relying on mathematical models of the software and the hardware involved.

Mathematical models can be used to generate timing estimations on source code level when the hardware is not yet fully accessible or the code is not yet ready to compile. In this thesis, the methods used to build timing models developed by WCET group in MDH have been assessed by evaluating the accuracy of the resulting timing models for a number of combinations of hardware architecture. Furthermore, the timing model identification is extended for various hardware platforms, like advanced architecture with cache and pipeline and also included floating-point instructions by selecting benchmarks that uses floating-points as well.

Place, publisher, year, edition, pages
2012. , 46 p.
Keyword [en]
Real-time systems, WCET analysis, simulation, Early timing analysis, SimpleScalar, SWEET, Linear timing models
National Category
Computer and Information Science
URN: urn:nbn:se:mdh:diva-15093OAI: diva2:543101
Subject / course
Computer Science
2012-06-14, Gamma, Mälardalen University, 721 23, Västerås, 16:15 (English)
Available from: 2012-08-10 Created: 2012-08-06 Last updated: 2012-08-10Bibliographically approved

Open Access in DiVA

Evaluation of a method for identifying timing models(2615 kB)315 downloads
File information
File name FULLTEXT01.pdfFile size 2615 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Kahsu, Lidia
By organisation
School of Innovation, Design and Engineering
Computer and Information Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 315 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 105 hits
ReferencesLink to record
Permanent link

Direct link