Database Statistics Emulation via Private Cache Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Emulating multiple target database systems on a test system is challenging, especially when many users are logged on, as existing methods disrupt the integrity of table statistics and lead to inefficient processing of requests.
Innovation Solution
A method that involves reading statistics from a target system's data dictionary, exporting them to a test system, and copying them to a cache for emulation, allowing for selective use based on user-defined levels of emulation, ensuring private statistics for each user and maintaining system integrity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If table statistics are extracted from the target database system's data dictionary and stored in the test database system's data dictionary, then the test system can emulate the target system, but when the test system is used to emulate more than one target database system with many users logged on, the integrity of table statistics is disrupted and processing efficiency deteriorates
Solution Approach 1:
The patent segments the statistics storage by creating separate private areas in the test database system's data dictionary for each user emulation session. Each private area stores statistics specific to one target database system, preventing cross-contamination between multiple emulations. This segmentation resolves the contradiction by maintaining statistics integrity (reliability) while supporting multiple emulations (adaptability).
Solution Approach 2:
The patent creates private copies of target database system statistics in each user's private area rather than sharing a single copy. When a user logs on to emulate a target system, the system copies the relevant statistics into that user's private area, ensuring that each user has an isolated, intact copy. This copying mechanism prevents statistics disruption while enabling multiple concurrent emulations.
2Ease of operation
If table statistics are stored in the test database system's data dictionary for emulation, then the optimizer can use them for query planning, but when multiple users are logged on emulating different target systems, the processing efficiency of requests deteriorates
Solution Approach 1:
The patent segments the data dictionary into public areas (shared by all users) and private areas (dedicated to each user's emulation session). Each private area contains statistics for a specific target database system emulation, allowing the optimizer to quickly access relevant statistics without scanning entire shared structures. This segmentation improves request processing efficiency while maintaining ease of operation for query optimization.
Solution Approach 2:
The patent performs preliminary action by pre-loading and storing target database system statistics into each user's private area before the user submits queries. This advance preparation ensures that when users submit requests, the optimizer can immediately access the required statistics without delay, thereby maintaining high productivity while preserving ease of operation.
3Device complexity
If the test system uses shared statistics in the data dictionary for multiple users, then device complexity is reduced, but the loss of information about specific target system statistics increases
Solution Approach 1:
The patent segments the data dictionary into public and private areas, where private areas store user-specific target system statistics. This segmentation preserves complete information about each target system for each user while maintaining a structured, manageable complexity through clear separation of shared and private data. The organized structure prevents information loss without creating excessive complexity.
Solution Approach 2:
The patent extracts specific target system statistics from the shared public area and places them into private areas for individual users. This extraction ensures that each user has access to the complete, unmodified statistics for their specific target system emulation, preventing information loss. The public area retains the base structure, so the overall system complexity does not increase significantly.
Data Source
AI summary
A method, system, and computer program are disclosed. Statistics are read from a data dictionary on a target system. The statistics include statistics for a table in a database in the target system. The statistics are exported to a test system separate from the target system. A command is received to use the statistics in emulation. In response the statistics are copied to a cache in the test system and the target system is emulated on the test system using the statistics in the test system's cache.