Intrusive Hash Table for Atomic Transaction Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Commercial database systems face performance degradation and increased latency when processing transactions due to the need for verifying transaction correctness and applying changes atomically, which can lead to decreased event throughput and higher memory usage.
Innovation Solution
The implementation of an intrusive hash table within a data management system for verifying and applying transaction operations in a two-pass process, allowing for atomicity and minimal performance and memory overheads, while collapsing operations to further improve performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional transaction verification and atomic change application methods are used, then transaction correctness is ensured, but event throughput decreases and latency increases
Solution Approach 1:
The patent segments the transaction processing into two distinct passes: a verification pass that checks transaction correctness without modifying data, and an application pass that applies committed transactions. This segmentation allows verification to proceed independently without blocking subsequent operations, thereby maintaining throughput while ensuring correctness.
Solution Approach 2:
The verification pass performs preliminary validation of transaction correctness before any data modifications occur. By pre-verifying transaction validity and identifying committed transactions ahead of time, the system avoids rework and enables efficient bulk application in the second pass, improving overall throughput.
2Reliability
If intermediate lookup values are cached for verification, then transaction correctness can be verified, but memory usage increases
Solution Approach 1:
The patent extracts the verification function from the data structure itself by using a separate intrusive hash table that stores only the necessary verification information (pointers to nodes and transaction status). This extraction eliminates the need to cache extensive intermediate lookup values while maintaining verification capability.
Solution Approach 2:
The intrusive hash table acts as an intermediary structure that provides verification information without requiring full caching of intermediate lookup values. It stores minimal pointer information that enables verification while keeping memory usage low, serving as a mediator between the transaction log and the actual data nodes.
3Reliability
If rollback procedures are implemented for transaction safety, then atomicity is maintained, but processing complexity and time increase
Solution Approach 1:
The verification pass performs preliminary identification of committed transactions before any data modifications. By determining which transactions are committed in advance, the system can apply them directly without needing rollback procedures, as there is no risk of partial application. This eliminates the time overhead associated with rollback mechanisms.
Solution Approach 2:
The patent extracts the rollback requirement entirely from the system by using a two-pass approach where verification and application are separated. Since committed transactions are identified and applied in a controlled manner during the application pass, there is no need for rollback capability, simplifying the system and reducing processing time.
Data Source
AI summary
Disclosed herein are system, method, and computer program product embodiments for utilizing an intrusive data hash table to support processing transactions. An embodiment operates by receiving a transaction comprising a plurality of operations. Responsive to the receiving, the transaction is verified by verifying the validity of the plurality of operations. Responsive to verifying the transaction, the transaction is applied to the intrusive hash table.