Mixed-Mode Transaction Architecture for Data Integrity and Scalability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in optimizing reliability, availability, and scalability, as Active/Active and Validation Architectures have limitations in data integrity and scalability respectively, necessitating a combined approach to enhance these attributes.
Innovation Solution
A mixed-mode architecture that combines Active/Active and DSR/TSR Validation Architectures, allowing transactions to be processed across multiple validation systems with replication, ensuring database synchronization and error detection, while accommodating varying transaction loads and node failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If Active/Active Architecture is used, then availability and scalability are improved, but data integrity and reliability deteriorate
Solution Approach 1:
The patent combines Active/Active Architecture and Validation Architecture into a hybrid system where validation nodes verify transactions before they are replicated across active nodes. This merging allows the system to maintain high availability through active replication while ensuring data integrity through validation checks, thus resolving the contradiction between reliability and productivity.
2Reliability
If Validation Architecture is used, then data integrity is improved, but scalability deteriorates
Solution Approach 1:
The patent segments the system into validation nodes and active replication nodes. Validation nodes perform integrity checks on transactions, while active nodes handle scalable transaction processing and replication. This segmentation allows validation to occur only when needed, enabling the system to maintain data integrity while scaling horizontally through additional active nodes.
3Reliability
If Active/Active replication is used, then availability is improved, but complexity of error detection worsens
Solution Approach 1:
The patent implements feedback mechanisms where validation nodes monitor transactions and provide verification feedback before commits. This feedback loop automatically detects errors in replicated data without requiring complex manual detection processes, thus maintaining high availability while simplifying error detection through systematic validation feedback.
Data Source
AI summary
A method is provided for performing transaction processing in a system that includes a transaction distributor and a plurality of nodes. Each of the nodes include a processor, an application executing in the processor, and an instance of a database used by the application. Each node executes transactions. The transaction distributor selects at least two of the nodes to receive the same request to process a transaction, and the processor at each of the at least two nodes executes the same request to process a transaction. Each of the at least two nodes replicates to one or more other nodes in the system the steps and operations of the transactions up until a transaction END step. The processor at the one or more other nodes compares the replicated steps and operations of the transactions that were received from each of the at least two nodes and performs a transaction END step for the replicated transaction at the one or more other nodes in the system by either executing a COMMIT when the result of the comparison indicates that the replicated steps and operations of the replicated transactions match each other, or executing an ABORT when the result of the comparison indicates that the replicated steps and operations of the replicated transactions do not match each other.