Determining the worst-case blocking sequences for distributed real-time programs

Abstract

For detecting timing errors in a distributed real-time program (distributed real-time debugging) it is essential to know all combinations of execution paths where a given path experiences a worst-case blocking time through communication (server calls) among the subprograms. In this paper we deal with the subproblem of blocking sequences of server call pairs between 2 paths. For determining their worst-case blocking time a polynomial algorithm had been found recently by D. Huizinga. We build on this result for constructing an efficient algorithm that computes all worst-case blocking sequences. Its correctness and practicality for distributed real-time debugging are discussed. Its extension to the general case of n+1 paths is outlined.<>