Multi-master Data Ownership Transfer via Hierarchical Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Database management systems face challenges in optimizing performance due to complex queries and large data volumes, particularly in handling ownership transfer of data partitions across multiple computing systems, which affects transaction time and query response.
Innovation Solution
A computer-implemented method for managing ownership of data partitions using a hierarchical partitioning structure replicated across multiple computing systems, allowing asynchronous updates and replication of data slices and nodes based on update frequency, enabling efficient data management and query processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If data partitions are replicated across multiple computing systems to improve availability and performance, then query response time is improved, but ownership transfer complexity increases
Solution Approach 1:
The database is divided into multiple data partitions that can be independently replicated and owned by different computing systems. Each partition can be selectively replicated to multiple systems, allowing the system to balance query response time requirements against ownership transfer complexity by managing only the necessary partitions across necessary systems.
Solution Approach 2:
The system implements dynamic ownership transfer where computing systems can assume ownership of data partitions based on usage patterns and workload requirements. This dynamic approach allows the system to adapt ownership assignments to current needs, improving query response time for frequently accessed data while reducing overall complexity by transferring ownership only when beneficial.
2Reliability
If frequent updates are performed on data partitions, then data freshness is improved, but replication overhead increases
Solution Approach 1:
The system dynamically adjusts replication behavior based on update frequency. When a computing system performs frequent updates on a data partition, the system automatically increases replication to keep other systems synchronized, ensuring data freshness. Conversely, when updates are infrequent, replication is reduced or paused, minimizing replication overhead and energy consumption.
Solution Approach 2:
The system monitors update frequencies and usage patterns of data partitions, using this feedback to dynamically adjust replication strategies. This feedback mechanism ensures that replication overhead is proportional to actual data freshness requirements, preventing unnecessary replication during low-activity periods while maintaining synchronization during high-update periods.
3Productivity
If data partitions are transferred between computing systems, then workload balancing is improved, but transaction time increases
Solution Approach 1:
The system performs dynamic ownership transfers of data partitions based on real-time workload analysis. When a computing system becomes overloaded or when another system has available capacity, the system automatically transfers ownership of appropriate data partitions to balance the workload. This dynamic approach improves overall productivity while minimizing transaction time by transferring only the necessary partitions during optimal moments.
Data Source
AI summary
A method, a system and a computer program product for executing management of ownership of data. An index in a plurality of indexes is selected. The index corresponds to a plurality of ranges of data values stored in a plurality of database slices of a database. The index further corresponds to a partitioning structure that includes a plurality of hierarchically arranged nodes. Each node corresponds to a range of data values stored in at least one database slice. The structure is replicated across a plurality of computing systems. A computing system executes an update to one or more ranges of data values. The system replicates at least one of a database slice including the updated ranges of data values and a node that includes the updated ranges of data values to another computing system for storage of a replicate of the updated ranges of data values.