databricks

Databricks workspace design requires configuring Unity Catalog and applying recommended environment separation patterns. This guidance provides layout configurations to isolate dev, test, and production environments securely within a single Lakehouse architecture.

Selecting a Databricks compute strategy involves evaluating Serverless, Classic, and Photon runtimes alongside DBR versions and autoscaling rules. This guidance enforces compute selection requirements to match pipeline workload demands while optimizing resource allocation.

Building a Medallion Architecture pipeline involves designing Bronze, Silver, and Gold data layers with DLT to automate data quality rules. This guidance structures your pipeline orchestration to enforce end-to-end data quality and Lakeflow operations.

Yes, you can use DABs and Terraform to automate Databricks CI/CD deployments and enforce cost controls. This approach provides end-to-end automated deployment pipelines integrated directly with your infrastructure provisioning and cost management rules.

Yes, this guidance covers integrating MLflow, Model Serving, Feature Store, and Mosaic AI into your Databricks architecture. It enforces requirements for the full ML lifecycle, from model training tracking to production serving within the Lakehouse.

Optimizing DBSQL performance involves tuning query execution and dashboard configurations within the Databricks Lakehouse. This guidance applies specific performance optimization requirements to reduce query latency and streamline dashboard execution.