Workflow scheduling using Neural Networks and Reinforcement Learning
Mikhail Melnik, Denis A. Nasonov · Procedia Computer Science · 2019
The development of information technologies entails a nonlinear growth of both volumes of data and the complexity of data processing itself. Scheduling is one of the main components for optimizing the operation of the computing system. Currently, there are a large number of scheduling algorithms. However, even in spite of existing hybrid schemes, there remains a need for a scheduling scheme that can quickly and efficiently solve a scheduling problem on a wide range of possible states of the computing environment, including the high heterogeneity of computational models and resources, and should have an ability to self-adapt and self-learn. At present, artificial intelligence and neural networks are the most popular methods for working with data and solving a wide range of problems, but they are not developed enough to solve the scheduling problem. Therefore, in this paper we propose a scheduling scheme based on Artificial Neural Networks and the principles of Reinforcement Learning.