Cell Conformance Testing with Respect to the Peak Cell Rate in ATM Networks

In this paper, we investigate how the concept of cell conformance testing developed within the CCITT could apply for connections with two types of tra c (namely with both CLP=0 and CLP=1 cells) in the Tra c Contract recommended by the CCITT (i.e., declaration of the peak cell rates of the CLP=0 and the aggregate CLP=0+1 cell streams). This issue is crucial in the sense that this Tra c Contract could be invalid if it were impossible to de ne an unambiguous conformance testing procedure. We show that most of the cell conformance testing mechanisms that could be naturally considered su er from an undesirable phenomenon, namely measurement phasing which jeopardizes their use. We then propose a mechanism for cell conformance testing which does not su er from any problem of measurement phasing and which can account for cell tagging. Along with this mechanism we propose a method of computing reference Cell Loss Ratios for connections with both CLP=0 and CLP=1 cells.