Feature Information Synchronization Across Training and Inference Databases
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in managing data consistency and convenience when separate databases are used for model training and inference, requiring separate data logic implementation and affecting model performance.
Innovation Solution
An electronic apparatus synchronizes data between databases with different characteristics, storing feature information based on quantity, attributes, and sources, and provides synchronized data to external entities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If separate databases are used for model training and inference, then data management flexibility is improved, but data consistency deteriorates
Solution Approach 1:
The patent introduces a feature information store as an intermediary component between the training database and inference database. This mediator receives feature information requests, determines whether to retrieve from training or inference database based on request characteristics, and returns appropriate results. This intermediary resolves the contradiction by enabling flexible data access patterns while maintaining data consistency through centralized management logic.
2Ease of manufacture
If separate databases are used for model training and inference, then database optimization for specific purposes is improved, but system complexity deteriorates
Solution Approach 1:
The feature information store implements a universal access interface that handles both training and inference database operations through a single system. The system determines at runtime whether to access the training database or inference database based on the feature information request characteristics, thereby providing multi-functional capability without requiring separate access paths for each database type.
3Manufacturing precision
If separate databases are used for model training and inference, then specialized data storage is improved, but development productivity deteriorates
Solution Approach 1:
The system implements self-service functionality by automatically determining which database (training or inference) should serve each feature information request based on the request characteristics. The feature information store autonomously makes routing decisions without requiring external intervention or complex coordination, thereby maintaining specialized data storage benefits while simplifying the development process.
Data Source
AI summary
Provided is an electronic apparatus and a method of managing feature information performed by the electronic apparatus, the method including the electronic apparatus receiving a request to store feature information in a database, based on at least some of a feature quantity, attributes of the feature information, a calculation method of the feature information and sources of the feature information, the electronic apparatus storing the feature information in one of a first database in which data for inference of a machine learning model is stored and a second database in which data for training the machine learning model is stored, and the electronic apparatus synchronizing the feature information from one of the first database and the second database in which the feature information is stored to the other database.