Statistical analysis of WCET for scheduling

Abstract
To perform a schedulability test, scheduling analysis relieson a known worst-case execution time (WCET).Thisvalue may be difficult to compute and may be overly pessimistic.This paper offers an alternative analysis basedon estimating a WCET from test data to within a specificlevel of probabilistic confidence.A method is presented forcalculating an estimate given statistical assumptions.Theimplications of the level of confidence on the likelihood ofschedulability are also presented.

This publication has 11 references indexed in Scilit: