A container image slimming method, device, equipment, medium and product
By combining decision trees and Hamming distance, container images are split into different types of image heaps, which solves the problem of low efficiency in container image slimming in existing technologies and achieves efficient deduplication and resource optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA MOBILE GROUP DESIGN INST
- Filing Date
- 2024-12-09
- Publication Date
- 2026-06-09
AI Technical Summary
Existing methods for slimming down container images are inefficient and cannot effectively handle a large number of large container images. Furthermore, these methods cannot be reused for slimming down other images, resulting in wasted resources and high system performance pressure.
A pre-built decision tree is used to split container images into different types of image stacks. By calculating the similarity of image layers of the same type, Hamming distance is used to determine the similarity and delete redundant files. Combined with federated learning to optimize the decision tree, efficient deduplication is achieved.
It improves the deduplication efficiency of large container images, reduces system performance pressure, and achieves efficient image layer slimming and resource utilization.
Smart Images
Figure CN122173195A_ABST