Cloud-Native Database Transaction Resume Across Node Switchover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud-native database systems face challenges in making database computing node switchover imperceptible to users and minimizing time resource overheads during node switchover, particularly in planned and unplanned scenarios.
Innovation Solution
Implement a cloud-native database system with a database state node to store transaction state information, allowing seamless continuation of transactions across computing nodes by storing and retrieving state information such as query plans, logs, and intermediate results, thereby enabling resumable processing at transaction or statement granularity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If database computing node switchover is performed in cloud-native database system, then system availability and scalability are improved, but time resource overheads increase and user experience deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-storing transaction state information (including query plans, intermediate results, and transaction logs) in external storage before node failure occurs. When switchover is needed, the new node can directly retrieve and resume these pre-stored states without needing to re-execute transactions or perform rollback/redo operations, significantly reducing time resource overheads during node switchover.
Solution Approach 2:
The patent introduces an intermediary component (external storage system or distributed file system) that mediates between the database computing nodes and transaction state information. This intermediary allows state information to be persisted independently of any specific computing node, enabling seamless switchover while maintaining fast access to transaction states without the overhead of traditional distributed transaction protocols.
2Adaptability or versatility
If database computing node switchover is performed, then system scalability and fault tolerance are improved, but switchover becomes perceptible to users and user experience worsens
Solution Approach 1:
The patent extracts the transaction state information from the computing node's memory and stores it in an external storage system. This separation allows the computing node to be swapped or scaled without affecting the persistence of transaction states. During switchover, users experience no interruption because the new node immediately retrieves the extracted state information and continues transaction processing transparently.
Solution Approach 2:
The patent maintains copies of transaction state information in external storage that can be quickly retrieved by any computing node. When a node fails or needs to be replaced, the new node obtains a copy of the transaction state from external storage and resumes processing, making the switchover imperceptible to users while enabling system scalability.
3Reliability
If traditional rollback or redo operations are used during node switchover, then transaction consistency is maintained, but processing time increases and productivity decreases
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing transaction state information (including query execution plans, intermediate results, and transaction logs) in external storage before node failure. During switchover, the new node directly retrieves these pre-computed states and continues transaction processing without needing to perform time-consuming rollback or redo operations, thereby maintaining transaction consistency while significantly improving processing efficiency.
Solution Approach 2:
The system enables self-service by automatically retrieving and resuming transaction states from external storage without requiring complex distributed transaction coordination protocols. The new computing node independently obtains the necessary state information and continues processing, eliminating the need for traditional rollback/redo mechanisms and improving productivity while maintaining consistency.
Data Source
Figure 1~4
Figure 5~6
Figure 7~8
AI summary
Embodiments of this application disclose a transaction processing method and a cloud-native database system, and belong to the field of database technologies. The method includes: A first database computing node responds to a transaction request of a user and starts to process a target transaction; the first database computing node stores state information of the target transaction into a database state node in a process of processing the target transaction; in the process in which the first database computing node processes the target transaction, the first database computing node suspends processing the target transaction; and a second database computing node obtains the state information of the target transaction from the database state node, and continues to process the target transaction based on the state information of the target transaction. In this application, after database computing node switchover is performed, resumable processing can be performed on a transaction, so that time resource overheads caused by the database computing node switchover are reduced, and the database computing node switchover is imperceptible to the user.