Constraint Data Statistics Objects for Database Partition Pruning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Database management systems (DBMS) face challenges in utilizing dynamic data integrity constraints due to their high cost of discovery and maintenance, which limits their application in improving query processing time and reducing memory consumption.

Innovation Solution

The use of traditional data statistics as dynamic data integrity constraints, specifically through constraint data statistics objects that imply and manage implied dynamic data integrity constraints, allowing for efficient creation and maintenance, and enabling dynamic partition pruning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If dynamic data integrity constraints are used to improve query processing time and reduce memory consumption, then query performance is improved, but the cost of discovery and maintenance increases significantly

Engineering Contradiction:
Improvequery processing timeVSAvoidcost of discovery and maintenance
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent uses data statistics objects as simplified copies or representations of dynamic data integrity constraints. These statistics objects capture essential information about data distribution and characteristics without requiring the full complexity of formal integrity constraints. The query optimizer uses these statistics copies to make pruning decisions, achieving performance benefits while avoiding the high maintenance costs of true dynamic constraints.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent employs data statistics objects that are inexpensive to create and maintain compared to formal dynamic integrity constraints. These statistics objects can be quickly generated, updated, and discarded as needed for query optimization. They serve as temporary, lightweight representations that enable partition pruning without the long-term maintenance burden of comprehensive integrity constraint systems.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

2Productivity

If data statistics objects are used as dynamic data integrity constraints, then query processing efficiency is improved, but ensuring consistency with database state becomes more challenging

Engineering Contradiction:
Improvequery processing efficiencyVSAvoidconsistency with database state
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements mechanisms to monitor and update data statistics objects based on actual database state changes. When data is inserted, updated, or deleted, the system provides feedback to refresh the relevant statistics objects. This feedback loop ensures that the statistics remain consistent with the current database state, maintaining reliability while enabling efficient query processing through partition pruning.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3401807B1Synopsis based advanced partition elimination
Publication Date: 2023.10.25 SAP SE
  • EP3401807B1 patent drawingFigure 1
  • EP3401807B1 patent drawingFigure 2
  • EP3401807B1 patent drawingFigure 3

AI summary

Disclosed herein are system, method, and computer program product embodiments for determining a constraint data statistics object is consistent. An embodiment operates by receiving a query for a database table. The embodiment determines a constraint data statistic object associated with a partition of the database table. The embodiment determines the constraint data statistics object is consistent with a data state of the partition of the database table based on consistency metadata of the constraint data statistics object. The embodiment processes the query for the database table based on the determining the constraint data statistics object is consistent with the data state of the partition of the database table.