Intelligent micro-service resource scheduling strategy based on load prediction

By predicting microservice load and building SLO violation scale, combined with multi-agent reinforcement learning algorithm, intelligent scheduling of microservice resources is realized, solving the problems of resource scheduling delay and resource waste in the existing technology, and ensuring the stable operation of system performance.

CN119718565AInactive Publication Date: 2025-03-28CHONGQING UNIV OF POSTS & TELECOMM
5 Cites 0 Cited by

Patent Information

Application Number
CN202411712535.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-27
Publication Date
2025-03-28
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing technology is difficult to schedule microservice cluster resources in a timely manner, resulting in inability to effectively respond when the load surges or drops sharply, resulting in insufficient resources or waste, affecting the stability of system performance.

Method used

By predicting the microservice load situation in advance, using graph neural network and SVM algorithm to build load prediction and SLO violation scale models, and combining multi-agent reinforcement learning algorithms to realize intelligent scheduling of microservice resources.

Benefits of technology

It realizes timely scheduling of microservice resources, avoids SLO violations and resource waste, ensures stable operation of system performance, and reduces resource consumption through global optimal resource configuration.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a micro-service resource intelligent scheduling strategy based on load prediction, and belongs to the technical field of micro-services. The method comprises the following steps: obtaining a topological structure of micro-services according to a micro-service calling relationship, and obtaining the request quantity, response time and CPU and memory resource use conditions of each micro-service deployment container; according to the collected request quantity of each container and the micro-service topological structure, predicting the request load of a future micro-service node by using a graph neural network; according to the obtained micro-service node request load, constructing a micro-service SLO violation model by using an SVM algorithm to obtain an SLO violation rate; and according to the obtained micro-service node request load condition and the SLO violation probability, a multi-agent-based reinforcement learning algorithm is utilized to obtain allocation configuration of micro-service node resources in the next time period. Compared with the prior art, the SLO guarantee is provided for the micro-service in the container cloud environment, the resource overhead can be reduced, and the overall performance of the system is improved.
Need to check novelty before this filing date? Find Prior Art