Skip to main content

Destinations overview

A destination is the terminal node of a pipeline — where Praxis collectors deliver telemetry after sources and processors are done. Praxis is designed so each pipeline can fan out to one or more destinations independently, so you can keep an existing SIEM while you adopt a new observability backend without rebuilding upstream collection.

Use the Integrations catalog to filter destinations by signal type, platform, and category. The pages here are the field-level reference behind each catalog entry.

How destinations are grouped​

CategoryExamples
SIEM / security data lakesGoogle SecOps, Splunk HEC, Microsoft Sentinel, AWS Security Lake, HPE OpsRamp
Observability backendsNew Relic, SigNoz, Dash0, generic OTLP
Object storageAWS S3, MinIO, Google Cloud Storage
Data warehousesSnowflake
DiagnosticsDebug — print to collector logs while you iterate.
SecOps data processingGoogle SecOps Data Processor Integration — pairs with the Google SecOps data processing pipeline flow.

Choosing a destination​

  • Single source of truth vs split — A pipeline can have multiple destinations, so common patterns include "send everything to S3 for retention, send a filtered subset to the SIEM, and a separate copy to the observability backend."
  • Format and signal compatibility — Most observability destinations accept OTLP natively. Security destinations often expect a specific log type or schema; the SecOps standardization processor and the catalog page for each destination show what to set.
  • Credentials live in integrations — Each destination references a saved integration (project, region, account, credentials). Update the integration in Praxis once and every pipeline using it picks up the change at next publish.