The use of hidden Markov models (HMMs) has found widespread use in many different areas. This chapter focuses on HMMs applied to the performance evaluation of computer systems and networks. After presenting a brief review of background material on HMMs, applications such as channel delay and loss characteristics, traffic modeling and workload generation are surveyed. The power of HMMs as predictors of performance metrics is also highlighted. We conclude by presenting a few features of the module of the Tangram-II performance evaluation tool that is targeted to HMMs.