Container Image Management via Metadata Mapping and Rule Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing container images in IT environments is challenging due to their large size, which consumes significant disk space and makes it difficult to identify and manage outdated, unsecured, or corrupted images, especially as the number of images increases.
Innovation Solution
A system that generates a mapping of unique identifiers and metadata for container images, validates this mapping against predefined rules to identify specific actions for images meeting certain conditions, and performs these actions, such as purging or relocating images based on size, access time, encryption status, and copyright information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If container images are stored in a repository, then software applications can be deployed and managed, but disk space consumption increases significantly and management becomes difficult
Solution Approach 1:
The patent segments container images into individual entries with unique identifiers in a mapping structure, separating metadata (JSON format containing size, creation time, labels) from the actual image data. This segmentation enables selective management and identification of individual images without handling the entire repository as a single unit, facilitating efficient storage management and retrieval.
2Adaptability or versatility
If the number of container images increases, then more software applications can be supported, but identification and management of outdated or insecure images becomes difficult
Solution Approach 1:
The patent implements a validation mechanism that generates feedback by comparing container image mappings against defined rules. The validation engine analyzes metadata (size, creation time, labels) and provides feedback on which images meet specific criteria (e.g., outdated, oversized, unencrypted), enabling automated identification and management actions without manual inspection of each image.
Solution Approach 2:
The patent replaces manual mechanical inspection of container images with an automated validation system that uses rule-based algorithms. Instead of manually examining each image file, the system substitutes mechanical review with computational validation that automatically compares image metadata against predefined rules, significantly reducing the difficulty of identifying outdated or insecure images.
3Extent of automation
If container images are validated against rules, then automated management actions can be performed, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by defining validation rules in advance before the validation process begins. Rules specifying criteria for image management actions (e.g., delete if older than X days, move if size exceeds Y GB) are established beforehand, allowing the validation engine to automatically execute appropriate actions without requiring complex real-time decision-making logic during validation.
Data Source
AI summary
Examples described relate to container images. An example method may include generating a mapping of respective unique identifiers of container images of a software container and respective metadata of the container images, validating the mapping against a rule to identify a unique identifier of a mapped container image that meets the rule, generating a target mapping that includes the unique identifier of the mapped container image that meets the rule and the action specified in the rule, and performing the action specified in the rule on the mapped container image corresponding to the unique identifier in the target mapping.