Skip to main content

Connect Data Store

A Data Destination is where DataBridge delivers your validated, high-quality events. This is typically a database where you run analytics, build dashboards, or train ML models.

You can check the list of supported data destination in our docs.

Connection Modes​

DataBridge supports two ways to connect to your database:

Cloud Mode (Default)​

DataBridge connects directly to your database over the network. Your connection credentials are stored securely in DataBridge.

  • Simplest setup - just provide host, port, username and password
  • Requires your database to be reachable from the DataBridge platform
  • Best for managed database services (RDS, Cloud SQL, Aiven, etc.)

DataBridge Agent Mode​

Coming Soon

Only data quality at rest monitoring and checks are supported via self-hosted agent at the moment. Data ingestion via Agent Mode is coming soon.

A DataBridge Agent running inside your infrastructure handles the connection. Credentials stay in the agent's local config file and never leave your network.

  • Database does not need to be publicly accessible
  • Credentials remain on your infrastructure
  • Best for private networks, VPCs and strict security requirements
  • See the Agent Guide for setup details

Creating Your First Destination​

Step 1: Navigate to Data Stores​

Go to Data Stores and click Connect New.

Step 2: Choose Destination Type​

Select your database: PostgreSQL, MySQL, or ClickHouse.

Step 3: Configure Connection​

Provide connection details. For example, for PostgreSQL:

Host: pg.example.com
Port: 5432
Database: analytics
Username: databridge_user
Password: ••••••••
Schema: events
SSL Mode: require

See the destination-specific docs for the full configuration reference:

Required permissions (PostgreSQL example):

GRANT USAGE ON SCHEMA events TO databridge_user;
GRANT CREATE ON SCHEMA events TO databridge_user;
GRANT SELECT, INSERT, UPDATE ON ALL TABLES IN SCHEMA events TO databridge_user;
ALTER DEFAULT PRIVILEGES IN SCHEMA events
  GRANT SELECT, INSERT, UPDATE ON TABLES TO databridge_user;

Step 4: Test Connection​

Click Test Connection to verify:

  • Credentials are correct
  • Network connectivity works
  • Required permissions are granted
  • Target schema/database exists

Step 5: Save Destination​

Click Save to store your configuration.

Schema-Driven Table Creation​

When a JSON Schema is registered for an event, DataBridge automatically creates tables with native column types. For example, in PostgreSQL:

CREATE TABLE events.purchase_completed (
  user_id TEXT,
  order_id TEXT,
  amount DOUBLE PRECISION,
  currency TEXT,
  items JSONB,
  timestamp TIMESTAMPTZ,
  ref_url_id TEXT,
  event_id TEXT
);

Type mappings vary by destination - see each destination's documentation for details.

If no schema is registered, the table falls back to a generic three-column layout (ref_url_id, event_id, payload).

Data Delivery Process​

  1. Buffering - Events are collected in batches to reduce the number of database operations
  2. Transformation - Optional Functions can enrich, filter, or modify events
  3. Loading - Batch insert into your database with automatic table creation and error handling

Multi-Destination Delivery​

You can send the same events to multiple destinations by creating multiple pipelines from the same source:

  • Production + Development databases
  • OLTP + OLAP (PostgreSQL for dashboards, ClickHouse for analytics)
  • Different transformations per destination

Each destination can have different transformations applied via pipeline configuration.

Best Practices​

  1. Use dedicated database users - Create separate credentials for DataBridge for easier auditing and access control
  2. Use a dedicated schema - Keep DataBridge tables organized in their own schema (e.g., databridge or events)
  3. Test with Blackhole first - Verify your pipeline logic using a Blackhole destination before connecting a real database
  4. Monitor delivery - Check pipeline metrics in the dashboard to track delivery success rates and latency

Next Steps​

  1. Configure a Pipeline to route events from source to destination
  2. Set up Transformations to enrich your data
  3. Enable Quality Monitoring for your database tables
  4. Configure Alerts to get notified about delivery issues