trino-lakehouse-platform-architect

What problem does it solve?

It helps you design a production-ready Trino-based modern data platform that cleanly separates compute from storage while providing a coherent lakehouse architecture and metadata/catalog strategy.

Core Features & Use Cases

• Trino query-layer architecture: define coordinator/worker topology and the principles for running Trino as a compute engine only.
• Catalog and metadata-driven integration: design Iceberg catalogs via Hive Metastore or Glue, and connect operational sources (e.g., PostgreSQL) and streaming sources (e.g., Kafka).
• Multi-layer lakehouse modeling: establish Bronze/Silver/Gold zones with appropriate Iceberg table settings, partitions, and performance-oriented options for analytics.
• Federated query across heterogeneous sources: support cross-catalog querying while guiding trade-offs and anti-patterns.
• Real-world example: model raw Kafka ingestions into Iceberg Bronze, transform and deduplicate into Silver using dbt, and publish curated aggregated facts into Gold for BI tools like Superset.

Quick Start

Ask your AI to draft a Trino lakehouse blueprint for a new platform on MinIO/S3 using Iceberg with Hive Metastore (or Glue), including coordinator/worker sizing guidance and Bronze/Silver/Gold table designs for Bronze raw events, Silver curated datasets, and Gold aggregated facts.