Traditional approaches for worst case execution time (WCET) analysis
aim at finding the absolute upper bound on the execution time.
For modern high performance processors with, for example, out-of-order execution,
these technique may produce estimates for the WCET which are pessimistic due to
the simplifications that may need to be made and due to the inherent variability
of the execution time. In addition, end to end measurements as used in industry
produce estimates of the execution time that potentially underestimate the real worst
case execution time.
We introduce the notion of probabilistic hard real-time system as a system
which has to meet all the deadlines but for which a (high) probabilistic
guarantee suffices.
pWCET
combines both measurement and analytical approaches into a model for computing
probabilistically bounds on the execution time of the worst case path of
sections of code.
The technique behind pWCET
is based on
combining (probabilistically) the worst case effects seen in individual blocks
to build the execution time model of the worst case path of the program (such
case may have not been observed in the measurements).
Update: The pWCET
technology is now available in a commercial tool: the RapiTime Worst Case Execution Time (WCET) analysis tool.
|
