This post details the final steps in designing a zero downtime migration solution with a focus on architecture transitions, rollback plans, and the overall migration process. It covers development tasks, such as updating batch applications, accounting code processing, historical data handling, and modifying queries for BigQuery. The importance of monitoring mutation counts in Cloud Spanner is also addressed, along with a summary of key takeaways and future work.

13m read timeFrom engineering.mercari.com
Post cover image
Table of contents
Development TasksRollback PlansSpanner Mutation Count EstimationMigration StepsFuture WorkKey TakeawaysConclusion

Sort: