Transaction Concurrency Control via Operation Rows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing transactional data management systems face challenges in maintaining data consistency and concurrency control, particularly in applications with high update rates and aggregated values, leading to write-write conflicts and rollbacks, especially when scaling out to multiple computing nodes.
Innovation Solution
A transactional semantic concurrency control system that manages databases using value rows and operation rows, where operation rows are substituted into value rows only after they are known to be visible, allowing for conflict-less data management and providing different isolation levels, including snapshot isolation and read committed, to reduce memory pressure and enhance scalability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If snapshot isolation is used to reduce read-write conflicts, then read concurrency is improved, but write-write conflicts still occur causing rollbacks
Solution Approach 1:
The patent segments the database rows into two distinct types: value rows (for reading) and operation rows (for writing). This segmentation allows read operations to access value rows while write operations modify operation rows, eliminating the write-write conflicts that cause rollbacks in traditional snapshot isolation while maintaining high read concurrency.
2Reliability
If traditional concurrency control is used to ensure data consistency, then data consistency is maintained, but write-write conflicts cause transaction rollbacks reducing productivity
Solution Approach 1:
The patent introduces operation rows as an intermediary layer between write operations and the actual data. Write operations first modify operation rows, which then trigger substitutions into value rows. This intermediary mechanism maintains data consistency through the substitution trigger while avoiding direct write-write conflicts that would cause rollbacks.
3Productivity
If multi-version concurrency control is used to enable snapshot isolation, then read concurrency is improved, but memory consumption increases due to storing multiple versions
Solution Approach 1:
The patent creates value rows as copies of the data state at specific points in time, rather than maintaining multiple versions of all data. When substitutions occur, new value rows are created only when necessary, and old value rows can be garbage collected. This copying approach enables snapshot isolation with controlled memory usage compared to full multi-version storage.
4Reliability
If operation rows are immediately substituted into value rows, then data consistency is maintained, but concurrency control efficiency is reduced due to conflict checking
Solution Approach 1:
The patent performs write operations on operation rows first, before any substitution into value rows occurs. This preliminary action on operation rows allows multiple write operations to proceed concurrently without conflict checking, and only triggers substitutions when appropriate, thereby improving concurrency control efficiency while maintaining data consistency.
Data Source
AI summary
The present invention refers to the field of transactional data management. The invention provides a transactional semantic concurrency control system for managing a database, wherein the system comprises a data store engine which is adapted to manage the database comprising value rows, and each value row comprising a primary key uniquely identifying each row, at least one semantic column and a commit timestamp; wherein the data store engine is further adapted to manage operation rows, which are rows comprising a primary key, at least one operation to be applied over at least one semantic column of a value row identified by the primary key, and a commit timestamp; store the value rows and the operation rows; generate an operation row when it receives an operation that inserts, updates, upserts, and/or deletes a row of the database with the commit timestamp of the operation; and apply the operations of at least one operation row when a trigger condition is satisfied.