G e state of research id parallel simulation now demands u m m : / w l t h a o f synchronization, communication, and event executions. models I t is evident that large-scale simulations involving many interacting logical processes should be a As researchers active in the study of synchronization focal point of such experimentation, as large-scale methods and load balancing, we have undertaken the simulations will benefit the most from parallelism This lmplenlentatlon of a Parallel simulation testbed, Yet realization raises a number of issues. Large-scale parallel Anofhe7 W~t z d o~~l~2 g N e f~~k S 2 n l~l n f~ (YAWNS) simulations must aggregate many logical processes onto which addresses these problenls. YAWNS 1s being each machine in a distributed memory architecture. This implemented on Our Intel ipsc/21 and Will Serve as the fact creates internal management problen1s---how does basis for our studies in the parallel simulation of timed Petri-nets, queueing networks, Monte Carlo physical s i ni u 1 at i on s, h i g hl y mu 1 t i processor what computing systems YAWNS provides a software layer between the LPs, and the communication network All communication and synchronization between processors i s initiated and controlled by individual version of YAWNS running on each processor, as shown in figure 1.YAWNS uses a both a new synchronous synchronization pro t o c o 1, and a n e w " rn u 1 t ior d e r " p r i or i t y-qu e u e

The state of research in parallel simulation now demands that we experiment with a multitude of simulation models. It is evident that large-scale simulations involving many interacting logical processes should be a focal point of such experimentation, as large-scale simulations will benefit the most from parallelism. This realization raises a number of issues. Large-scale parallel simulations must aggregate many logical processes onto each machine in a distributed memory architecture. This fact creates internal management problems---how does one synchronize in such a setting? How does one efficiently find and manage the simulation workload? If we are to experiment with multiple models, what underlying functions can we extract to program once, and use many times? This paper describes YAWNS, Yet Another Windowing Network Simulator, for dealing with these problems.