A model that estimates how fast a system can process requests and how many it can handle per unit time.