Zero-Outage Database Reorganization via Shadow Copy Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Database reorganization in partitioned databases typically requires taking the database offline, leading to visible outages and potential losses for businesses, especially those in banking and financial sectors, due to the time-consuming process of renaming hundreds or thousands of data sets.
Innovation Solution
A method and system for reorganizing a database by creating a shadow copy, taking a partition offline while maintaining a logical view online, replacing data set names in the schema without causing transactions to fail, and quiescing the database to allow for quick updates without renaming data sets, thus avoiding the need for exclusive access and reducing downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If database reorganization is performed by renaming data sets, then the database can be reorganized, but the database must be taken offline causing visible outages
Solution Approach 1:
The database is divided into partitions, each with its own data sets. The reorganization process operates on individual partitions independently, allowing other partitions to remain online and accessible. This segmentation enables partial reorganization without taking the entire database offline.
Solution Approach 2:
A shadow copy of the database schema is created before the actual reorganization. The shadow schema allows preparation and testing of reorganization changes in advance, enabling the actual reorganization to proceed with minimal disruption to the live database.
2Ease of manufacture
If data set names are renamed serially, then the database schema can be updated, but the process takes minutes causing significant downtime
Solution Approach 1:
The shadow schema is created and populated with updated data set names before the actual reorganization begins. This preliminary preparation allows the schema update to be performed in bulk rather than serially, dramatically reducing the time required for the actual reorganization operation.
Solution Approach 2:
A shadow copy of the database schema is created with the updated data set names. This shadow schema serves as a template that can be applied to the actual database schema in a single operation, eliminating the need for serial renaming of thousands of data sets.
3Ease of manufacture
If exclusive access to catalog files is required for renaming, then data set names can be changed, but applications cannot be online during the rename
Solution Approach 1:
The catalog file access is segmented by partition. Each partition has its own catalog entries that can be updated independently. This allows the reorganization to proceed on one partition while other partitions remain accessible to applications, eliminating the need for exclusive access to the entire catalog file.
Data Source
AI summary
Methods and systems enable a database reorganization to occur without a database outage. In one aspect, the method includes creating a shadow copy of a database, the shadow having at least one partition associated with a plurality of first data sets and the database having at least a first partition associated with a plurality of second data sets and a second partition associated with a plurality of third data sets. The method also includes reorganizing the at least one partition of the shadow, taking the first partition offline, replacing the plurality of second data sets with the plurality of first data sets in a schema, and restarting the first partition.