Distributed Database Data Object Grouping for Transaction Response Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed databases face performance issues due to increased response times caused by the need for client interface nodes to communicate with multiple database nodes during transactions, especially in scenarios with dynamic and evolving data usage conditions, where manual optimization of data distribution is difficult and static solutions fail to maintain performance.
Innovation Solution
A method of grouping data objects involved in transactions at a single database node, with dynamic allocation and management of data object groups to reduce the need for distributed transactions, using a data distribution manager to coordinate data object grouping, copying, and distribution across nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of stationary object
If data objects are distributed across multiple database nodes to increase storage capacity, then the storage capacity of the distributed database is improved, but the response time of database transactions worsens due to the need to communicate with multiple nodes
Solution Approach 1:
The patent segments data objects into groups that are co-located on the same database node. By dividing the data storage into logical groups (data object groups) and assigning each group to a specific node, the system maintains scalable storage capacity while reducing transaction complexity. When data objects are grouped together, transactions involving multiple data objects can be executed on a single node, eliminating the need for inter-node communication and thus reducing response time.
Solution Approach 2:
The patent introduces a data distribution manager that acts as a coordinating entity to manage data object grouping and distribution. This manager counterbalances the performance degradation caused by distributed transactions by intelligently organizing data objects into groups that minimize cross-node access. The manager tracks which data objects are accessed together and ensures they are stored on the same node, effectively counteracting the negative impact of data distribution on transaction performance.
2Productivity
If manual configuration is used to optimize data distribution for specific applications, then transaction performance is improved, but the complexity of system configuration and maintenance increases
Solution Approach 1:
The patent implements a self-service mechanism where the data distribution manager automatically monitors data access patterns and dynamically adjusts data object grouping without manual intervention. The system observes which data objects are frequently accessed together during transactions and automatically groups them on the same node. This self-adjusting capability maintains optimal transaction performance while eliminating the need for complex manual configuration and ongoing maintenance of data distribution settings.
Solution Approach 2:
The data distribution manager incorporates feedback loops that continuously monitor transaction patterns and data access behavior. Based on this feedback, the manager dynamically reorganizes data object groups to optimize for observed access patterns. This feedback-driven approach enables the system to automatically adapt to changing workloads and data usage patterns, maintaining high transaction performance without requiring manual reconfiguration or prior knowledge of application requirements.
3Volume of stationary object
If the number of database nodes is increased to handle larger data volumes, then the storage capacity is improved, but the likelihood of distributed transactions increases, worsening system performance
Solution Approach 1:
The patent merges data objects that are frequently accessed together into the same data object group, regardless of which physical node currently stores them. By combining related data objects into logical groups and ensuring each group resides on a single node, the system allows transactions to access all required data objects from one node even as the number of physical nodes increases. This merging strategy enables the distributed database to scale storage capacity while maintaining transaction throughput comparable to single-node operations.
Data Source
AI summary
According to a first aspect of the present invention there is provided a method of operating a distributed database comprising a plurality of database nodes each storing a plurality of data objects. The method comprises, when executing a database transaction that requires data operations be performed on a plurality of data objects that are distributed between two or more of the database nodes, grouping the data objects that are involved in the database transaction at one of the plurality of database nodes.