Dynamic assignment based on a probabilistic matching: Application to server-container assignment

Abstract Matching algorithm has been used in broad domains of assignment problems because of its simplicity and efficiency. We also studied a probabilistic matching focusing on the application to digital systems including IoT. This note reports a modification of the probabilistic matching aiming application to the dynamic assignment of computational resources and devices This note discusses the dynamic assignment between servers and containers. Container technology is widely used because of its simplicity and efficiency compared with the virtual machines. This note reports the merit and demerit of using the dynamic matching with the simulation of the scenario using CPU resource and memory as dynamic parameters reflecting the changing environment.