Numerical simulation of rotor stator interaction is crucial for turbo machinery improvement since it allows the optimization of blade path design In addition the numerical simulation of rotor stator interaction makes it possible to predict the location and variation of hot spots on blades which is extremely useful in the turbine thermal design However simulation of rotor stator interaction is notorious for being computationally expensive To reduce the turnaround time and cost MFLOP a parallel code based on the message passing interface libraries was developed This code runs on symmetric multi processors Silicon Graphics Challenge and massively parallel processors Cray T E The parallel code was used to compute the rotor stator interaction in two turbine configurations A super linear speedup of was obtained on a CPU Silicon Graphics Challenge for a geometry distributed over ten processors