Database Table Storage with Dynamic Regular-LOB Space Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Storing large objects (LOBs) in database systems is challenging due to their large data sizes, which can exceed the maximum size of regular data types, leading to inefficient storage and data movement overheads.
Innovation Solution
A method for storing tables with a mix of regular and large object columns by allocating separate physical spaces for each, comparing recorded data volumes, and dynamically assigning data between these spaces based on volume, minimizing data movement through swapping or deallocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If separate physical spaces are allocated for regular and large object columns, then storage efficiency is improved, but device complexity increases
Solution Approach 1:
The patent divides the table storage into two separate physical spaces: regular table space for small columns and large object table space for LOB columns. This segmentation allows each space to be optimized independently, improving overall storage efficiency while managing complexity through clear separation of concerns.
Solution Approach 2:
The patent introduces a new dimension to storage management by creating a hierarchical structure with regular table space at one level and large object table space at another level, connected through locator columns. This dimensional approach resolves the contradiction by organizing data across multiple storage layers rather than in a single flat structure.
2Loss of time
If data is stored in separate physical spaces, then data movement overhead is reduced, but storage management complexity increases
Solution Approach 1:
The patent performs preliminary actions by pre-allocating separate physical spaces for regular and large object columns before data insertion. Locator columns are pre-configured with placeholders that will reference the large object storage locations. This preliminary organization eliminates the need for data movement during subsequent operations, reducing time loss while managing complexity through upfront planning.
Solution Approach 2:
The patent introduces locator columns as intermediaries between the regular table space and large object table space. These locators act as reference pointers that connect small column data with large object data without requiring physical data movement. This intermediary mechanism reduces data movement overhead while providing a manageable structure for storage operations.
3Adaptability or versatility
If dynamic allocation between physical spaces is implemented, then storage adaptability is improved, but operational complexity increases
Solution Approach 1:
The patent implements dynamic storage allocation by allowing the system to determine at load time whether to use the first or second physical space based on the actual data characteristics. The storage structure can adapt its configuration dynamically during table loading, optimizing for the specific data patterns encountered rather than being fixed in advance.
Solution Approach 2:
The patent changes storage parameters dynamically by switching between different physical space configurations based on data volume and type analysis. The system can adjust whether to allocate primarily to regular table space or large object table space depending on the specific table being loaded, providing adaptability while managing operational complexity through parameter-based decision making.
Data Source
AI summary
The present disclosure relates to a method for storing a table in a database system. The table comprises a first set of one or more columns of first data type and a second set of one or more columns of second data type. The method comprises: storing the first set of columns and a set of locator columns in a first regular table space. The second set of columns may be stored in a second large object table space. In case a first recorded data volume is higher than or equal to a second recorded data volume, data of the second large object table space may be assigned to a first large object table space. In case the first recorded data volume is smaller than the second recorded data volume, data of the first regular table space may be assigned to a second regular table space.