Parallel replica dynamics was proposed by A.F. Voter as a tool for accelerating molecular dynamics simulations characterized by a sequence of infrequent, but rapid, transitions from one state to another. An example would be the migration of a defect through a crystal. Parallel replica dynamics accelerates this by simulating many replicas simultaneously, concatenating the simulation times of the realizations as though it were a single long trajectory. This motivates important questions: Is parallel replica dynamics algorithm doing what we hope? For what systems will it be useful? How do we implement it efficiently? In this talk, I will thoroughly describe the algorithm and report on progress towards rigorous justification. Open questions and related problems will also be discussed.