Using online worst-case execution time analysis and alternative tasks in real time systems
As embedded hardware becomes more powerful, it allows for more complex real
time systems running tasks with highly dynamic execution times. This dynamicity
makes the already formidable task of producing accurate WCET analysis even
more di?cult. Since the variation in execution time depends on task input and
the state of the system, it is postulated that a more accurate estimate for the
WCET can be found online with knowledge about the task parameters.
This thesis will explore the concept of online execution time analysis and its
potential utilization. Line detection in images through Hough line transform is
found to be a relevant application whose execution time can be estimated by
the contrast of the input image. A system for scheduling tasks utilizing their
online WCET estimate is then discussed. It dynamically checks for potential
deadline misses and degrades tasks, either by running a more e?cient alternative
task instead or by aborting the task, until timely execution is guaranteed. An
experiment is presented, demonstrating a higher throughput of tasks with online
WCET estimation. Finally, the work on a framework for more precise simulations
and experiments is presented.
Place, publisher, year, edition, pages
Institutt for teknisk kybernetikk , 2014. , 82 p.
IdentifiersURN: urn:nbn:no:ntnu:diva-26100Local ID: ntnudaim:10746OAI: oai:DiVA.org:ntnu-26100DiVA: diva2:744134
Hendseth, Sverre, Førsteamanuensis