Core
Hydrolix uses several "Core" components to operate the system, including the Database bucket and the Catalog:
Database Bucket⚓︎
Database buckets use the name convention hdxcli-xxxxxx. Each Database bucket contains:
- platform configuration
- table and database configurations
- platform tunables
- logs
- data files encoded in the HDX file format
Hydrolix stores database files as partitions. Each partition corresponds to a time period, and contains both raw data and indexes.
For example, an AWS deployed service uses the following directory structure:
Catalog⚓︎
The Catalog is a database instance used to manage state information for the stateless components. The Catalog contains:
- metadata for the data partitions stored within the Database bucket
- information on the jobs and tasks executed as part of ingestion
You can query the Catalog using the reserved view for each table, accessible through the suffix #.catalog:
Other Components⚓︎
Hydrolix also uses the following minor components:
| Component | Description | Scale to 0 |
|---|---|---|
| Keycloak | Manages Role-Based Access Control (RBAC) and access control for services. | No |
| Operator | Kubernetes Operator that manages Hydrolix scale and infrastructure. | No |
| Validator | Service for testing Hydrolix Transforms that translate data from input to the internal schema. | Yes |
| Version | Endpoint that displays the current running version of the Hydrolix platform. | No |
| Vector | Deployed on each Node, used to collect Logs from each service. | No |