This repo aims to provide example setups that can be used to build towards streamlined Model Ops workflows. This initial version provides a docker-compose setup that provides the following services:
- trino server - unified SQL query interface
- iceberg service - high performance columnar table storage capability
- hive metastore collects metadata from connected data sources for heuristic query performance optimization
- minio server - cloud provider agnostic object storage system with S3 compatible interface
All of these services can be accessed via the trino interface. An example notebook showing output of an example connection and initial queries is provided for demonstration.