🪴 Ogród Wiedzy

GraphQL Subscriptions

Properties2
tagsgraphql, subscriptions, websockets, real-time
aliasesGraphQL Subscriptions, Subscriptions

2 min. czytania

🎯 Definition

GraphQL Subscriptions are a feature of the GraphQL specification that allows servers to push real-time, event-driven data updates to connected clients. Unlike Queries (fetch once) and Mutations (write once), Subscriptions establish a long-lived, bidirectional connection—typically over WebSockets—to push notifications immediately when specific database or application events occur.

🔑 Key Takeaways

  • Real-time Push: Clients receive data automatically without polling the server.
  • Protocol Underlay: While standard GraphQL queries and mutations use HTTP POST, subscriptions require a protocol that supports persistent connections, such as WebSockets (e.g. ws:// or wss:// protocols) or Server-Sent Events (SSE).
  • Stateful Backend: Subscriptions increase server memory usage because the server must maintain active connections for every client and track event pub/sub (publish-subscribe) networks (e.g., via Redis Pub/Sub).

📚 Detailed Explanation

GraphQL Subscriptions are designed for pushing updates about incremental modifications to large, dynamic datasets.

The client registers a subscription query with the server:

subscription OnRecordCreated {
  recordCreated {
    id
    name
    status
  }
}

The subscription flow works as follows:

  1. Handshake: The client initiates a WebSocket connection to the server.
  2. Registration: The client sends the subscription query document. The server registers this connection to a specific event stream (e.g., “record creation”).
  3. Event Trigger: A separate mutation (or backend database change) publishes an event.
  4. Data Push: The server resolves the subscription query fields for that event and pushes the JSON payload to the listening client over the WebSocket.

In metadata platforms like Ataccama ONE, subscriptions are utilized to dynamically update the catalog web UI (e.g., showing a progress bar of a running profiling job or alerting when a new catalog item has been created).

💡 Use Cases & Examples

  • Real-time DQ Alerting: A browser dashboard listening for Data Quality alerts. When a validation rule fails, the backend triggers an event, and the server pushes the DQ failure notification directly to the UI.
  • Monitoring Pipeline Progress: Pushing incremental load counts from an ETL workflow directly to a console.

📌 Sources

Graf

Ostatnie notatki