The original authors (see ibid., vol.7, no.3, p.424 (1989)) describe some results from a simulation study of ENET II and Ethernet. ENET II is a method, proposed by the commenter (Comput. Networks ISDN Syst., vol.9, no.3, p.209-14, Mar. 1985), to implement several variants of well-known collisions-resolution protocols in an asynchronous environment in a manner upwardly compatible with Ethernet. The original authors concluded that the ENET II protocol is inferior to the standard Ethernet. The commenter compares the assumptions in the simulation whose results were expressed in the original paper to their assumptions and makes several points about the analysis they described. The thrust of the comment is not to negate the results but to make the reader aware of the subtle differences in the representation of the results that make it difficult to conclude that ENET II is either inferior or superior to Ethernet.>

Two protocols for the data-link level of local area networks, namely, the original Ethernet protocol and the EnetII protocol, are compared under varying load conditions. The two protocols were implemented on a testbed network. The protocols were found to have similar throughput and delay characteristics under normal network conditions, while Enet III achieved higher throughput and greater stability under extreme overload conditions. The predicted delay advantage for Enet II was not observed in these measurements due to implementation compromises, but Enet II was observed to achieve its goal of greatly reducing the variance of the packet queuing delay. It was also noted that, contrary to previously reported results, network utilization is sensitive to the number of traffic-producing hosts, for five or more hosts.>

Basic graph models of processes, such as Petri nets, have usually omitted the concept of time as a parameter. Time has been added to the Petri net model in two ways. The timed Petri net (TPN) uses a fixed number of discrete time intervals. The stochastic Petri net (SPN) uses an exponentially distributed random variable. In this paper, a discrete time stochastic Petri model is described. These discrete time SPN's fill the gap between TPN and normal SPN. However, the use of discrete time complicates the SPN model in that more than one transition may fire at a time step. Finally, an example of a live and bounded Petri net which has nonempty, disjoint, recurrent subsets of markings is given.

Several options are specified in the X.3 standard for the selection of packet construction time. Some simple queueing theoretic models are constructed to study two of the three alternative criterion, fixed packet size and end of packet character.

An isomorphism between the behavior of Petri nets with exponentially distributed transition rates and Markov processes is presented. In particular, k-bounded Petri nets are isomorphic to finite Markov processes and can be solved by standard techniques if k is not too large. As a practical example, we solve for the steady state average message delay and throughput on a communication link when the alternating bit protocol is used for error recovery.