Queueing System with Two Phases of Service and Service Rate Degradation

In the paper, a queueing system with an unlimited number of servers and two phases service degradation in rate is studied. The problem emerges cloud nodes, where there contention for hardware resources including computational such as CPU cores. node, we have limited cores that should execute potentially processes (requests) parallel. our model, term “server” means process allocated node execution. So, “servers” but their individual performances decrease because CPUs switch between them during We consider executed life cycle reflects periods different activity process; e.g., first phase, may require intensive usage low second one. Our model distinguishes using parameters well influence on node. analytical methods are proposed: exact solving local balance equation asymptotic analysis global equations. Formulas stationary probability distribution customers obtained both cases. Several numerical examples provided illustrate some properties applicability results.