System and method for data redistribution in a database

By recording snapshots and splitting data in a massively parallel processing database, the problem of transaction failures caused by faults during data redistribution is solved, ensuring data consistency and integrity, and improving the reliability and efficiency of the system.

CN112965951BActive Publication Date: 2025-10-24HUAWEI TECH CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202110172742.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2016-08-03
Filing Date
2017-07-06
Publication Date
2025-10-24
Estimated Expiration
2037-07-06

AI Technical Summary

Technical Problem

In massively parallel processing databases, during data redistribution, the failure of a single device or node can cause the entire transaction to fail, resulting in wasted processing time and resources. Furthermore, existing technologies struggle to maintain data consistency and integrity when failures occur.

Method used

By recording snapshots of job data, splitting the data into a primary data portion and a redistributed data portion, and accumulating changes in a temporary table, data migration and merging are achieved. This ensures that data consistency can be restored in the event of a successful or failed migration, allowing faulty nodes to continue processing.

Benefits of technology

This ensures that data consistency and integrity are maintained even in the event of a failure during data redistribution, avoiding redundant operations and resource waste, and improving the reliability and efficiency of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112965951B_ABST
    Figure CN112965951B_ABST
Patent Text Reader

Abstract

A method for redistributing job data in a first data node (DN) to at least one additional DN in a massively parallel processing (MPP) database (DB) is provided. The method includes recording a snapshot of the job data, creating a first data portion in the first DN and a redistribution data portion in the first DN, collecting changes to a copy of the job data stored in a temporary table, and initiating migration of the redistribution data portion to the at least one additional DN.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • System and method for massively parallel processor database

    WO2016116052A1