The use of discrete-state, continuous- time Markov processes (MP, for short) has been shown useful for the modeling and evaluation of Fault -Tolerant Computing Systems. The efficient application of Fault-Tolerant Computing Systems. The efficient application of this approach to complex systems depends largely on the availability of powerful software tools that provide “friendly” utilities to specify and run the models, and that can deal with large size MPs. Both aspects have been considered in the design and implementation of a software tool called METFAC. This paper discusses the framework solutions on which the tool is based and presents some performance metrics. In relation to the framework solutions, the following aspects are considered: generative specification of the models by production rules, and sparse analytical evaluation of the measures.