Online migration scheduling method, device, equipment and system based on container integration

Abstract

The invention discloses an online migration scheduling method, device, equipment and system based on container integration. Comprising the steps of obtaining load values corresponding to a plurality of nodes, where the load values corresponding to the nodes are used for indicating load conditions of the nodes corresponding to N continuous time windows from a current time window to the front; inputting the load value of each node into an overload prediction model, and obtaining a prediction result of each node, where the prediction result is used for indicating whether the node is overloaded in the next time window of the current time window, the overload prediction model comprises a convolutional neural network layer and a gating circulation unit layer, and a training sample of the overload prediction model comprises supervised data; and migrating the to-be-migrated container of the overload node to the target node, where the nodes comprise the target node, and the prediction result of the overload node indicates that the overload node is overloaded in the next time window of the current time window. Scheduling and allocation of resources in the container cloud environment are realized, and the utilization rate of the resources and the energy efficiency are improved.

Description

technical field [0001] The invention belongs to the technical field of computers, and in particular relates to an online migration scheduling method, device, equipment and system based on container integration. Background technique [0002] With the rapid development of computer technology, the high scalability of container technology has not only been generally recognized in the industry, but also as the core technology of cloud computing, it supports the comprehensive development and popularization of IoT, 5G and other technologies. [0003] However, as the scale of the container cloud cluster continues to expand, on the one hand, the resource load of the cluster in this operating environment is changing all the time. Within a certain period of time, the rapid growth of container resource requests will inevitably lead to an increase in the resource load of the node. When the upper limit is exceeded, it will cause the node to be overloaded, which will cause a decline in the...

AI Technical Summary

Problems solved by technology

Within a certain period of time, the rapid growth of container resource requests will inevitably lead to an increase in the resource load of the node. When the upper limit is exceeded, it will cause the node to be overloaded, which will cause a decline in the performance of external services and cannot be timely and effective. Complete user needs

Images