Application Placement in Fog Computing based on Deep Q-Learning Algorithm

Fog computing paradigm has been suggested as a complement to cloud computing to support large volumes of heterogeneous data generated. The existence of this new processing paradigm also created challenges. Application placement in Fog computing is one of this challenges. The modern application is made of non-integrated components that can be controlled and delivered to you. Deep Q-learning algorithm, which is a special kind of reinforcement learning algorithm, is a good solution for embedding applications in the fog computing environment. We apply deep Q-learning algorithm to solve application placement problem. The simulation results show that the response time of application placement using deep Q-learning method is 44.39% reduced compared to the non-learning queue algorithm method.

کلیدواژه ها:

Fog computing ، application placement ، machine learning ، deep reinforcement learning ، Q-learning