Task Scheduling in Cloud Using Deep Reinforcement Learning

Cloud computing is an emerging technology used in many applications such as data analysis, storage, and Internet of Things (IoT). Due to the increasing number users cloud IoT devices that are being integrated with cloud, amount generated by these ceaselessly. Managing this over no longer easy task. It almost impossible move all datacenters, will lead excessive bandwidth usage, latency, cost, energy consumption. This makes it evident allocating resources users’ tasks essential quality feature computing. because provides customers or high Quality Service (QoS) best response time, also respects established Level Agreement. Therefore, there a great importance efficient utilization for which optimal strategy task scheduling required. paper focuses on problem cloud-based aims minimize computational cost under resource deadline constraints. Towards end, we propose clipped double deep Q-learning algorithm utilizing target network experience relay techniques, using reinforcement learning approach.