GCP Database Engineer case studies

GCP Database Engineer · Put your knowledge to work in context.

The Professional Cloud Database Engineer is a Google Cloud certification aimed at database professionals who design, build, manage, and troubleshoot the cloud databases that applications rely on for storing and retrieving data. Google describes the target candidate as someone with roughly two years of Google Cloud experience and five years of overall database and IT experience, comfortable turning business and technical requirements into scalable, cost-effective database solutions. The GCP Database Engineer case studies are 5 cases and about 5,546 words.

Seat Inventory With a Global Consistency SLA and a Sub-150 ms Write Target

TicketNova sells concert seats in 14 countries. Seat inventory lives in Cloud SQL for PostgreSQL Enterprise Plus: primary in us-central1, cross-region read replicas in europe-west1 and asia-east1.

1,077 words, about 5 min 2

CDC Pipeline Lagging 40 Minutes and Producing Duplicate Revenue Rows

A retailer runs Cloud SQL for MySQL 8.0 (16 vCPU, 600 GB) for orders, Datastream for CDC into BigQuery, and Bigtable for cart and session serving. Finance reports on orders in BigQuery.

1,067 words, about 5 min 3

Oracle 19c to Google Cloud With a Four-Hour Cutover Window

A logistics company runs Oracle 19c on-premises: 6.4 TB of data, 340 tables, 1,800 PL/SQL packages, 2,100 scheduled DBMS SCHEDULER jobs, and 180 synonyms pointing at a second Oracle instance that will not be migrated. The largest table is a 2.9 TB shipment event table partitioned by month.

1,121 words, about 6 min 4

400 GKE Pods, 6,000 Idle Connections, and an RTO Nobody Has Tested

A SaaS product runs on GKE Autopilot with 400 pods across three zones in us-central1. Each pod opens 15 direct connections to a Cloud SQL for PostgreSQL Enterprise Plus instance (32 vCPU, 208 GB) and holds them idle for the process lifetime.

1,136 words, about 6 min 5

Bigtable Hotspot: 180,000 Writes/s Into One Row-Key Range

An observability vendor ingests device metrics into Cloud Bigtable. Row keys are deviceID#reverseTimestamp , a design chosen so the newest metrics sort first for a single device.

1,145 words, about 6 min