Distributed Graph Metadata Version Filtering for Lower Management Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed graph databases, managing graph metadata becomes inefficient and costly due to the need to maintain and manage multiple data storage service versions, leading to increased management load for the engine analysis service.
Innovation Solution
The method involves starting data storage services in a debugging mode to obtain actual service version configuration data, creating graph metadata without active services, and updating it with node metadata when services are operational, ensuring the engine analysis service only manages supported versions, thereby reducing unnecessary load and costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the engine analysis service maintains and manages all data storage service versions supported by the distributed graph database, then the system supports multiple versions for flexibility and compatibility, but the management load and costs increase while efficiency decreases
Solution Approach 1:
The patent extracts and removes unsupported data storage service versions from the graph metadata management scope. The engine analysis service only manages graph metadata for currently supported data storage service versions, eliminating the need to maintain and manage all historical versions, thereby reducing management load while preserving adaptability for active versions
Solution Approach 2:
The patent implements dynamic version support where the engine analysis service adapts its management scope based on currently supported data storage service versions. The system dynamically determines which versions to manage by checking compatibility during deployment, allowing flexible adaptation to changing version requirements without maintaining fixed management of all historical versions
2Adaptability or versatility
If the engine analysis service manages all data storage service versions, then version flexibility is maintained, but graph metadata management costs increase
Solution Approach 1:
The patent extracts and removes unsupported data storage service versions from management scope. By filtering out versions that are no longer supported, the system reduces the quantity of graph metadata that needs to be managed, directly lowering management costs while preserving flexibility for currently supported versions
Solution Approach 2:
The patent changes the parameter of version management scope from managing all historical versions to managing only currently supported versions. This parameter change reduces the volume of metadata management work and associated costs while maintaining the necessary flexibility for active version deployment
3Adaptability or versatility
If the engine analysis service manages all data storage service versions, then system compatibility is ensured, but graph metadata management efficiency decreases
Solution Approach 1:
The patent extracts and removes unsupported versions from the management scope. By focusing management efforts only on currently supported data storage service versions, the system improves management efficiency while maintaining compatibility for active versions through targeted metadata management
Solution Approach 2:
The patent applies partial action by managing only the subset of data storage service versions that are currently supported, rather than managing all historical versions. This partial management approach improves efficiency by concentrating resources on active versions while maintaining sufficient compatibility for deployment purposes
Data Source
AI summary
Storage service version configuration data is obtained at a time when a data storage node cluster starts a data storage service based on a debugging mode. Based on the storage service version configuration data and data storage node cluster topology information, graph metadata of the distributed graph database is created at a time when no data storage service is provided, where the graph metadata includes a storage service version configuration field populated with the storage service version configuration data and a node metadata field of each data storage node. Node metadata information of the data storage node is obtained at a time when the data storage node cluster starts the data storage service based on a normal mode. The obtained node metadata information of the data storage node is filed into a corresponding node metadata field of the graph metadata to obtain managed graph metadata.


