An enterprise supply chain platform processes high-throughput freight tracking telemetry using a microservices application deployed on Google Kubernetes Engine (GKE), backed by Cloud SQL for PostgreSQL. The development team is preparing to release a major version update that includes non-backward-compatible database schema changes. The business requires zero downtime during release and immediate rollback capabilities if application errors increase. Which TWO implementation steps must the cloud architect mandate to safely execute this release?
- Apply an expand-contract database migration pattern by introducing additive schema changes (new columns/tables) and maintaining dual-write support before initiating application deployment.Cevap
- Configure GKE Native Ingress traffic splitting to incrementally route canary traffic to the new microservice revision while monitoring SLO error budgets.Cevap
- CExecute a blue-green application cutover while executing a single transactional DDL script that drops and renames active database tables.
- DProvision a complete standalone duplicate GKE cluster in the target region immediately prior to deployment without requesting regional compute quota adjustments.
Cevap
The cloud architect must implement an expand-contract database migration pattern to ensure backward compatibility and configure GKE Native Ingress traffic splitting to safely run a canary release.
To achieve zero downtime when database schemas change, the expand-contract (parallel change) pattern must be applied so the database supports both application versions concurrently. Paired with GKE Ingress traffic splitting (canary deployment), traffic can be gradually routed to the new release while monitoring error budgets.
Adım Adım Çözüm
Anahtar Kavram
Release Management and Deployment Strategies (Blue-Green, Canary, Rolling)