Cloud platform resource dynamic scheduling method based on deep learning

Abstract

A cloud platform resource dynamic scheduling method based on deep learning comprises the following steps: obtaining time series data from monitoring information and log files of a system, and taking the time series data as a training data set of a model; constructing a workload prediction model; constructing a response time prediction model; and constructing a scheduling strategy model according to the constructed workload prediction model and the response time prediction model: performing joint debugging on the scheduling strategy model, optimizing model parameters, forming a stable cloud platform resource dynamic scheduling model based on deep learning, and realizing scheduling of cloud platform resources. Based on server-oriented scheduling, the deep learning model is used for continuous training, a stable and effective model is finally obtained, automatic scheduling of cloud platform resources can be achieved, a large number of cloud resources can be saved in actual production, and good economic benefits are achieved.

Description

technical field [0001] The invention relates to the field of automatic operation and maintenance, in particular to a method for dynamic scheduling of cloud platform resources based on deep learning. Background technique [0002] Cloud computing is an emerging leading information technology. Cloud computing is the distribution of computing tasks on the "cloud". Automated management through dedicated software enables users to access computing power, storage space, and information services on demand. Users can focus on themselves business without having to consider complex technical details, which helps to improve efficiency, reduce costs and technological innovation. [0003] According to existing surveys, the number of cloud computing users will continue to grow at least until 2026. The demand for cloud computing in the commercial and civilian fields is increasing. The growing number of users also brings pressure on data scale and service operations. expansion and increase. ...

AI Technical Summary

Problems solved by technology

However, due to the diversity of applications on the cloud computing platform and the dynamic change of user load, the problem of load imbalance between nodes may occur, some nodes are insufficiently loaded and resources are insufficient, while other nodes are idle and idle, which greatly affects the Overall Performance of Cloud Computing System

In addition, as the scale of cloud computing resources expands, manual or manual management of resource clusters becomes very impractical

Images