Guide

Databricks Endpoint

Databricks endpoints connect Gateway to Databricks SQL warehouses for governed lakehouse querying and analytics workflows.

Endpoint Kind

Use databricks as the endpoint kind when creating this endpoint through the API or dashboard.

What Gateway Uses This Endpoint For

Run governed SQL against Databricks SQL warehouses.
Expose curated lakehouse data to AI Orchestration and approved tools.
Apply RBAC, audit, and request analytics to warehouse access.
Use Databricks alongside warehouse and database endpoints in workflows.

Basic Endpoint Shape

json

{
  "endpoint": "databricks-warehouse",
  "kind": "databricks",
  "config": {
    "read_conn": {
      "host": "adb-1234567890.7.azuredatabricks.net",
      "http_path": "/sql/1.0/warehouses/abc123",
      "access_token": "secret"
    }
  }
}

Create it with the endpoint API:

bash

curl -sS -X POST "$EDEN/endpoints" \
  -H "$AUTH_HEADER" \
  -H "Content-Type: application/json" \
  -d '{
  "endpoint": "databricks-warehouse",
  "kind": "databricks",
  "config": {
    "read_conn": {
      "host": "adb-1234567890.7.azuredatabricks.net",
      "http_path": "/sql/1.0/warehouses/abc123",
      "access_token": "secret"
    }
  }
}'

Configuration Fields

Field	Purpose
`host`	Databricks workspace host
`http_path`	SQL warehouse HTTP path
`access_token`	Databricks personal access token or service principal token

Operating Notes

Prefer service principals or short-lived tokens for production access.
Scope warehouse permissions to the catalogs and schemas Gateway needs.
Track warehouse size and concurrency because Gateway can centralize many users through one path.