How Decentralized Data Products are Changing GEMA from Inside Out
summary
GEMA adopts a lakehouse and data mesh approach, builds a self‑service platform on Databricks on GCP with Terraform, and claims rapid adoption — over 250 data products and ~450 users onboarded, with 90 appointed owners/stewards and 2.5 PB/month analytics throughput. Teams automate deployment with repo YAMLs, expose bronze/silver/gold layers as consumable products, deliver member services and 11 AI projects, cut legacy costs ~€1M, and report measurable productivity gains.
AI-written summary · about
chapters (12)
- 0:00Decentralized data at GEMA
- 2:37Traffic laws for data governance
- 4:01Centralized teams challenges
- 7:18User-driven governance need
- 8:44Four-quart governance model
- 10:27Medallion pattern for products
- 13:20Data ownership roles
- 15:52Enterprise data platform
- 19:49250 data products growth
- 25:51Scaling automation strategies
- 32:09Learnings from journey
- 34:42Q&A
related talks
54mData Mesh in Practice
36mData Mesh: Kicking Off a Mentality Change in an Organisation
42mFour Years of Data Mesh in Practice
39mData Mesh in the wild
30mFrom Data Lake Entanglement to Data Mesh Decoupling: Scaling a Self-Service Data Platform
33mData: The Land DevOps Forgot
20m3-phased approach to delivering a lakehouse with data mesh
58mManifesto for a DevOps-like Culture Shift in Data & Analytics