Fault tolerant, large scale multi-server systems require an optimum number of repairmen for maximising performability. However, performability evaluation of such systems is difficult due to the state space explosion problem. In this paper, a simple and flexible approximate technique capable of overcoming state space explosion problem in computing the performability of large Markov models is presented. For validation of results, simulation has been used. It is shown that, this approach can handle large state space. The proposed method allows analysing the breakdown repair behaviour of large multi server systems. An optimisation study is presented together with numerical results showing the relationship of the number of servers and the number of repairmen for optimum performance.