Avoid Scalability Problems With Data Migration to New Releases: Two Patterns That Reduce Workload Release-to-Release

Software that's based on a database, and is regularly upgraded with new releases, can generate myriad combinations of migration paths. As the number of releases grows, the increasing number of migration paths to be supported becomes time-consuming and expensive. This paper learns how two migration patterns can help one avoid release-to-release scalability problems in data migration. The patterns, based on serial data transformation and hub-and-spoke structure, let one build an automated, reusable, and flexible migration framework in which the migration paths from all previous releases to a new release can be easily implemented.