Federated Database Metadata Extraction for Accelerator Containers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Federated database systems lack actual data integration among disparate databases, and the implementation of additional protocols like 2PC or 3PC is computationally expensive and can block large parts of data during synchronization.
Innovation Solution
A method involving a database management system and an accelerator system that maintains metadata information, detects trigger events, creates incomplete metadata for absent data containers, and dynamically identifies structural features to create complete metadata, allowing for efficient data container creation and processing without the need for additional protocols.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If additional protocols like 2PC or 3PC are implemented for data synchronization in federated database systems, then data consistency between constituent databases is improved, but computational overhead increases and data blocking occurs during synchronization
Solution Approach 1:
The patent extracts the metadata management functionality from the central DBMS and places it in the accelerator system. The accelerator system independently creates and manages metadata for data containers, eliminating the need for complex synchronization protocols between the DBMS and accelerator. This extraction resolves the contradiction by maintaining data consistency through local metadata management rather than through expensive inter-system protocols.
Solution Approach 2:
The patent implements preliminary action by having the accelerator system pre-create incomplete metadata structures before actual data processing occurs. The DBMS creates placeholder metadata entries that describe the structure of data containers that will be created in the accelerator. This preliminary metadata creation allows the system to prepare for data synchronization without immediately incurring the computational overhead of full synchronization protocols.
2Reliability
If additional protocols like 2PC or 3PC are implemented for data synchronization, then data consistency between constituent databases is improved, but data blocking occurs during synchronization
Solution Approach 1:
By extracting metadata management to the accelerator system, the patent eliminates the data blocking issue. The accelerator system independently manages its own metadata without needing to block or synchronize with the DBMS during metadata operations. This allows the DBMS to continue processing queries while the accelerator independently manages its data container metadata, maintaining both data consistency and continuous data availability.
Solution Approach 2:
The preliminary creation of incomplete metadata structures allows the system to reserve names and namespaces for future data containers without immediately blocking data access. Other data containers can continue to be accessed and processed while the metadata framework is being prepared, thus maintaining productivity while setting up the structure for future data consistency.
3Manufacturing precision
If complete metadata information is maintained centrally in the DBMS for all data containers, then data structure control is improved, but network traffic increases when synchronizing metadata between DBMS and accelerator
Solution Approach 1:
The patent segments metadata management into two parts: the DBMS maintains a catalogue with metadata for data containers it directly manages, while the accelerator system independently creates and manages metadata for data containers it creates. This segmentation eliminates the need for continuous metadata synchronization over the network, reducing network traffic while maintaining data structure control through the centralized catalogue that the DBMS can query when needed.
Solution Approach 2:
The patent introduces an intermediary mechanism where the accelerator system dynamically identifies structural features of data containers and communicates only this specific information to the DBMS. Instead of transferring complete metadata structures, the accelerator acts as an intermediary that selectively shares only the structural feature information needed by the DBMS to maintain its catalogue, significantly reducing network traffic while preserving data structure control.
4Adaptability or versatility
If the accelerator system dynamically creates data containers with unknown structures, then adaptability is improved, but metadata management complexity increases
Solution Approach 1:
The patent applies preliminary action by having the DBMS create incomplete metadata structures with placeholder information (such as names and namespaces) before the accelerator system creates the actual data containers. This preliminary framework provides a structured approach to dynamic creation, reducing metadata management complexity by establishing a predictable pattern for how dynamic data containers will be integrated into the system's metadata catalogue.
Solution Approach 2:
The patent implements feedback by having the accelerator system dynamically identify structural features of created data containers and communicate this information back to the DBMS. The DBMS uses this feedback to update its catalogue with complete metadata information. This feedback loop allows the system to maintain centralized metadata control while accommodating dynamic data container creation, as the accelerator's structural information automatically enriches the central catalogue without requiring complex manual metadata management.
Data Source
AI summary
A distributed database system (or federated database system) comprises a constituent database management system (DMBS), an accelerator system and a single interface. The accelerator system creates a data container in accordance with structural features identified by the accelerator system and stores data in the data container. The constituent database system supplements incomplete metadata associated with the data container to create complete metadata descriptive of the namespace and structure of the data container created by the accelerator system.