How It Works

Tracker

StreamProcessor is tracker agnostic - use existing trackers (Google Analytics, Snowplow, etc.) or build your own, emit data as json or query string from any data source (web, mobile, server-side, webhooks, etc.).

Collector

Serverless collection of data in real-time, protected by CORS, API keys or IAM and persisted in Google PubSub.

Streamer

Serialization and validation of data according to associated schemas (avro). Annotated fields are pseudonymized and the data can be written to PubSub (avro or json), BigQuery and Cloud Storage. BigQuery tables are created and patched automatically by the Streamer and with field descriptions according to the schema descriptions.

Registrator

Registrator is a serverless and RESTful schema registry that stores schemas in pubsub schemas and enables schema evoultion in a controlled fashion.

Processor

Execute streaming SQL queries and write the results in real-time to PubSub or Firestore to activate your data in better customer experiences or internal efficiency.