CMDB Dormant Data Deletion via Dependency Graph
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional configuration management databases (CMDBs) face issues with storage consumption and migration times due to overfilled data, leading to prolonged response times, and deleting outdated configuration items can cause inconsistencies if not managed properly.
Innovation Solution
A method is introduced where configuration items in a CMDB are labeled as deletion candidates based on age thresholds, ensuring only those items with all direct and indirect parent nodes also labeled are deleted, preventing inconsistencies and reducing storage consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If configuration items are retained in the CMDB indefinitely, then data completeness is maintained, but storage consumption increases and migration times prolong
Solution Approach 1:
The system performs preliminary identification of outdated configuration items by comparing their last update time against a threshold before deletion. This preliminary action ensures that only truly outdated items (those not updated within the threshold period) are candidates for removal, maintaining data completeness for recently updated items while preparing to reduce storage consumption by removing obsolete data.
2Quantity of substance
If configuration items are deleted without checking dependencies, then storage space is reduced quickly, but database inconsistencies occur
Solution Approach 1:
Before deleting configuration items, the system performs a preliminary check of dependency relationships using a directed graph data structure. It identifies all parent nodes of the candidate for deletion and verifies that all of them are also marked for deletion. This preliminary dependency verification ensures that only safe deletions are executed, preventing database inconsistencies while still enabling storage space reduction.
3Productivity
If the CMDB is cleaned regularly, then response time improves, but data loss may occur if deletion criteria are too aggressive
Solution Approach 1:
The system uses a configurable time threshold parameter to control the deletion behavior. By adjusting this threshold, users can balance between cleaning frequency and data retention requirements. The system marks configuration items for deletion only if their last update time is older than the threshold, providing a controllable mechanism that improves response time through regular cleaning while preventing unnecessary data loss through parameter tuning.
Data Source
AI summary
Embodiments of the present invention provide automated systems and methods for selectively finding and deleting dormant data (out of data, inactive data) in a configuration management database (CMDB), or in a database which has similar characteristics. The method includes using a directed graph where each node of the directed graph represents a configuration item, and traversing the directed graph in order to determine inactive or dormant data, and all of the objects which depend on that data. Configuration items are deleted only when all of the other configuration items from which it depends contain dormant data.