Self-hosted · v1.0

Realtime data collection,
without the database overhead.

Stream events from any client over WebSocket. Store them compactly as CSV. Query, monitor and export — all from one lightweight portal.

Live ingestion stream
1,284 events / min
14:28:01py-client-01{ sensor: "temp", value: 22.4 }42 B
14:28:01py-client-02{ user: "u_8291", action: "click" }38 B
14:28:02py-client-01{ sensor: "humidity", value: 61 }40 B
14:28:02py-client-03{ order_id: "A-2041", status: "shipped" }44 B

Built for ingestion at scale.

01 — Features
⚡

Realtime over WebSocket

Durable Socket.IO connections. No polling, no repeated API calls.

▤

CSV-first storage

Compact per-day rotation. Minimal database load, easy to export.

⚿

JWT authentication

Every SDK client verified before a single byte is accepted.

◈

Redis buffer & cache

Batch writes to disk. Pub/Sub ready for horizontal scale-out.

◐

Live dashboard

Watch events arrive in real time. Filter by client, date or payload.

⇩

One-click export

Download filtered ranges as CSV or JSON. Streaming, not blocking.

Drop in the SDK. Start streaming.

02 — Python SDK
# pip install datastack-sdk from datastack import Client client = Client( url="wss://your-host:4000", api_key="sk_live_..." ) client.connect() while running: client.send({ "sensor": "temp", "value": 22.4 })
How it works

A single line to send data.

The SDK holds a durable WebSocket connection, auto-reconnects on network drops, and batches events under backpressure. The server verifies your JWT, buffers events in Redis, and appends them to a per-client CSV file.

Ready to self-host?

Deploy with Docker in minutes. Full source, no vendor lock-in.