Heterogeneous service, computing power and power coupling mapping method for computing power nodes

By constructing a unified measurement system for heterogeneous business needs and computing resources, the measurement and scheduling problems in heterogeneous computing resource management have been solved, achieving efficient and green computing resource scheduling and improving resource utilization and energy efficiency.

CN122387683APending Publication Date: 2026-07-14SICHUAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SICHUAN UNIV
Filing Date
2026-05-06
Publication Date
2026-07-14

AI Technical Summary

Technical Problem

Existing technologies lack a unified measurement and scheduling method, making it difficult to effectively manage and schedule heterogeneous computing resources, resulting in low resource utilization and low energy efficiency.

Method used

Construct a unified measurement system for heterogeneous business needs and computing resources. By constructing business demand vectors and computing power supply vectors, perform computing power resource assessment and service quality assessment, and migrate tasks when resource constraints cannot be met, thereby realizing the coupling mapping of heterogeneous businesses, computing power and power.

Benefits of technology

It enables precise quantification and efficient scheduling of heterogeneous computing resources, improving resource utilization and system energy efficiency, and reducing energy consumption while meeting business needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122387683A_ABST
    Figure CN122387683A_ABST
Patent Text Reader

Abstract

The application discloses a heterogeneous service, computing power and power coupling mapping method for computing power nodes, relates to computing power resource data modeling technology, and comprises the following steps: in a heterogeneous service demand unified dimension layer, mapping the received heterogeneous service request to the demand amount of computing resources, storage resources and communication resources, and constructing a service demand vector; in a heterogeneous computing power resource unified dimension layer, uniformly describing the heterogeneous server computing power supply capacity in the computing resource, storage resource and communication resource dimensions, and constructing a computing power supply vector; in a computing power service-resource matching layer, performing computing power resource evaluation and service quality evaluation according to the service demand vector and the computing power supply vector; if the current target computing power node cannot simultaneously satisfy the resource supply hard constraint and the delay constraint, performing task migration logic; if the condition is satisfied, forwarding the service task to the node for execution. The application solves the problem that the prior art lacks a unified measurement and scheduling method.
Need to check novelty before this filing date? Find Prior Art