The London Stock Exchange Group (LSEG) Capital Markets Business unit has improved its Blue/Green software deployment methodology by utilizing continuous logical database replication. Previously, manual backup and restore processes were time-consuming and involved shutting down the production database to maintain consistency between Blue and Green environments. Logical replication has significantly reduced deployment time by minimizing downtime of the Blue environment.
Blue/Green deployment involves switching between production and staging environments to implement changes effectively, minimizing downtime and enabling easy rollbacks if needed. This approach has allowed LSEG to reduce operational costs and improve availability by avoiding production downtime during deployments.
By implementing logical replication, along with Amazon Aurora PostgreSQL-Compatible Edition and Amazon RDS for PostgreSQL, LSEG has streamlined its deployment process and achieved more agility in deploying new features. This also supports internal testing of application changes and eliminates the need for frequent backup and restore cycles, providing a significant advantage.
The post details a step-by-step process for implementing a Blue/Green deployment architecture using Aurora PostgreSQL Global Database. The solution overview involves serving internal users through Amazon Route 53, Application Load Balancer, and a Kubernetes cluster, interacting with an Aurora global database. Key steps include configuring cluster parameters, creating publications and subscriptions, verifying replication lag, and performing pre-cutover validation.
Considerations such as stopping write traffic to the Blue environment, schema changes, replication, handling triggers, and initial hydration of the Green environment are outlined. Post-cutover steps involve stopping the Blue cluster with a final backup and cleaning up unnecessary resources. The use of non-admin accounts, securing replication user credentials, managing schema changes, and validating changes before cutover are emphasized.
The post concludes by highlighting the benefits of Blue/Green deployment for Aurora PostgreSQL Global Database instances, including reduced deployment time, minimal downtime, cost savings, and the ability to test new application versions before production deployment. Feedback on the outlined steps is encouraged, as authors Ionut Bruma, Jonathan Kerr, Inthu Anandakumar, and Richard Chester welcome insights from users who try out the deployment process. Their expertise in Solutions Architecture, Technical Account Management, and Financial Services contributes to the comprehensive guide on Blue/Green deployments in the capital markets division.
Article Source
https://aws.amazon.com/blogs/database/how-london-stock-exchange-group-optimised-blue-green-deployments-for-amazon-aurora-postgresql-global-database/