Graph Database Segmentation for Large Data Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current graph databases struggle to efficiently store larger data due to limitations in handling data sizes exceeding a certain threshold.
Innovation Solution
A data storage method that involves splitting data exceeding a predetermined threshold into data blocks, determining block identifiers based on sequence, and storing data identifiers and block identifiers in a graph database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is stored in a graph database without splitting, then storage simplicity is maintained, but the database cannot support larger data sizes
Solution Approach 1:
The patent divides large data into multiple data blocks with sequential identifiers (block_id, offset, length). Each block is stored separately in the graph database, allowing the system to handle data sizes that exceed single-record limitations while maintaining structured access through the segmentation framework.
2Quantity of substance
If data is split into blocks, then larger data sizes can be stored, but query complexity increases
Solution Approach 1:
The patent pre-establishes a mapping relationship between the original data identifier and block identifiers during the storage phase. This preliminary action creates an index structure that enables efficient retrieval of block sequences during queries, avoiding the need for complex search operations on split data.
3Reliability
If data is stored as complete records, then data integrity is maintained, but resumable transfers are not supported
Solution Approach 1:
By dividing data into transferable blocks with metadata (offset, length, block_id), the system enables resumable transfers where interrupted downloads can be resumed from the next block. Data integrity is maintained through sequential reconstruction of blocks into the complete original data record.
Data Source
AI summary
This specification discloses data storage methods, apparatuses, devices, and storage media. One method comprises receiving data to be stored, wherein the data comprise data corresponding to a plurality of fields, determining whether a size of the data exceeds a predetermined threshold, and in response to determining that the size of the data exceeds the predetermined threshold: splitting a portion of the data to obtain data blocks, determining block identifiers of the data blocks based on a sequence of the data blocks, storing a data identifier of the data and the block identifiers of the data blocks that correspond to each other, and storing the block identifiers and the data blocks that correspond to each other.