Date Added: Jun 2012
In designing Switched Local Area Networks (Switched LANs), there is the inherent need of being able to compute all the end-to-end delays and hence, the average end-to-end delays of such LANs. The widely held notion in the literature is that, the ability to perform this computation is contingent on the formulation of an origin-destination traffic matrix of the network with respect to the hosts that are attached to it. The authors have shown in a previous paper that this notion does not seem to be correct. In this paper, they explain a methodology for enumerating all the end-to-end delays of any switched LAN and hence, of determining the average (or average of maximum) end-to-end delay of any LAN.