Approximation Algorithms for Feasibility Analysis in Real - Time Static - Priority Systems ∗

Feasibility tests determine whether it is possible for a given real-time system to always meet all of its timing constraints on a specified processing platform. Current feasibility tests for the uniprocessor static-priority scheduling of sporadic task systems run in pseudo-polynomial time. We present a fully polynomial-time approximation scheme (FPTAS) for feasibility analysis in static-priority systems. This test is an approximation in the sense that that there is a quantifiable trade-off between the fraction of the processor’s capacity that must be left unused, and the running time of the feasibility test.