This paper presents our work on the simulation of distributed memory parallel computers. We design a distributed simulator that takes as input an application written for a MIMD computer and run it on a workstations cluster with just a recompilation of the code. The hardware of the target machine is simulated so that the behavior of the application is identical to a native run on the simulated computer with virtual timings and trace file. Moreover, our analysis sets up the conditions required to achieve a good speedup as a function of the number of simulation hosts, the network latency and the granularity of the application.