Trino Catalogs
Trino catalogs define the connection to underlying data sources. Each catalog maps to a specific connector and data source configuration. The MATIH platform manages catalogs as part of the data infrastructure, with tenant-level access controls determining which catalogs are visible to each tenant.
Default Catalogs
| Catalog | Connector | Description |
|---|---|---|
delta | Delta Lake | Primary analytical data store |
iceberg | Apache Iceberg | Table format for large-scale analytics |
system | System | Trino system metadata and information |
memory | Memory | In-memory tables for temporary data |
Catalog Configuration
Catalogs are configured via properties files mounted into the Trino coordinator:
# delta.properties
connector.name=delta_lake
hive.metastore.uri=thrift://hive-metastore:9083
delta.metadata.cache-ttl=5m
delta.max-partitions-per-writer=100
hive.s3.path-style-access=true
hive.s3.endpoint=http://minio:9000
hive.s3.aws-access-key=${S3_ACCESS_KEY}
hive.s3.aws-secret-key=${S3_SECRET_KEY}Tenant Catalog Access
Each tenant has a configured set of accessible catalogs. The Query Engine validates catalog access before query execution:
| Tier | Accessible Catalogs |
|---|---|
| FREE | delta, memory |
| STARTER | delta, memory |
| PROFESSIONAL | delta, iceberg, memory |
| ENTERPRISE | All catalogs including custom connectors |
Custom Catalogs
Enterprise tenants can request custom catalog configurations for connecting to their own data sources. Custom catalogs are deployed as additional properties files in the Trino configuration.
| Step | Description |
|---|---|
| 1 | Tenant requests custom catalog through the catalog service |
| 2 | Platform validates connector availability and credentials |
| 3 | Catalog properties file is generated and deployed |
| 4 | Trino coordinator is notified to load the new catalog |
Catalog Metadata
The Query Engine exposes catalog metadata through the data catalog service, which queries Trino's information_schema for:
- Available schemas within each catalog
- Table listings per schema
- Column definitions and data types
- Table statistics for cost estimation