Database Synchronization via Segmented Transaction Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data synchronization methods face challenges in balancing response performance and data consistency, particularly when prioritizing synchronization of data relevant to the currently executed application or login user, leading to inflexibility in controlling data granularity and availability.
Innovation Solution
A data synchronization method that performs first synchronization processing on data referred to by the currently executed application in volatile memory and second synchronization processing on data processed in the same transaction in the synchronization destination database, ensuring high-speed synchronization of only the data needed while maintaining data consistency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data synchronization is performed for all data in the master database to the device database, then data consistency is improved, but response performance and data availability deteriorate due to the time-consuming synchronization process
Solution Approach 1:
The patent segments the synchronization process into two distinct phases: first synchronization processing for data currently referred to by the executed application, and second synchronization processing for the remaining data in the same transaction. This segmentation allows the system to prioritize synchronization of critical data while deferring less critical data, thereby improving response performance without completely sacrificing data consistency.
Solution Approach 2:
The patent introduces dynamic prioritization by identifying data that is currently referred to by the executed application and giving it higher priority for synchronization. The system dynamically adjusts synchronization priorities based on real-time application state and data usage, allowing critical data to be synchronized first while less critical data follows, thus balancing consistency and performance.
2Productivity
If data synchronization is performed preferentially for data referred to by the currently executed application, then response performance is improved, but data consistency may deteriorate due to partial synchronization
Solution Approach 1:
The patent performs preliminary identification of data that will be needed by the executed application before synchronization begins. By预先 identifying and prioritizing this data for first synchronization processing, the system ensures that critical data is synchronized first, improving response performance while maintaining a clear path to subsequent synchronization of remaining data to ensure consistency.
Solution Approach 2:
The patent ensures continuous synchronization action by implementing both first and second synchronization processing within the same transaction. The first synchronization processing handles critical data immediately, while the second synchronization processing continues to synchronize remaining data, ensuring that the synchronization action is continuous and ultimately achieves complete data consistency without interruption.
3Reliability
If the granularity of data synchronization is controlled at the transaction level, then data consistency is maintained, but flexibility in selecting specific data to synchronize deteriorates
Solution Approach 1:
The patent further segments the transaction-level data into two categories: data referred to by the currently executed application and other data in the same transaction. This additional segmentation within the transaction boundary allows the system to apply different synchronization priorities to different portions of the same transaction, thereby gaining flexibility while maintaining the transactional consistency guarantee.
Solution Approach 2:
The patent applies different synchronization priorities to different portions of the same transaction based on local characteristics. Data referred to by the executed application receives first synchronization processing with higher priority, while other data in the same transaction receives second synchronization processing. This local differentiation allows flexibility in data selection while maintaining overall transaction consistency.
Data Source
AI summary
In a method for writing data having been subjected to transaction processing in a synchronization source database to data in a synchronization destination database, a synchronization processing management unit of the synchronization destination database requests, as synchronization target data, first data that is a part of the data having been subjected to the transaction processing from the synchronization source database, receives at least the first data from the synchronization source database, performs first synchronization processing on the first data in a cache memory, and performs second synchronization processing on second data, in the synchronization destination database, which has been processed in a same transaction as a transaction in which the first data has been processed in the synchronization source database.