How it works

Stored first. Then delivered.

Every durable message is stored and numbered before anyone sees it. Here is the path it takes, and what happens when a client drops and comes back.

How a message travels through Notifiers Your app and its clients ask the regional endpoint, a load balancer, which relay host to use, then connect to that relay host over a WebSocket. Relay hosts store each publish in the durable database. Echo reads each stored message and fans it out to the relay hosts, which deliver it to clients. Swengle's management for the region runs the hosts. THE REGION YOU PICK Your app and its clients your server, browsers and apps Regional endpoint load balancer: lookup, HTTP API YOUR CLUSTER (OR A SHARED ONE) Relay hosts connections, auth, limits, publishing Echo fan-out Database stored, numbered, replayable Management starts, scales and replaces hosts 1 HTTPS 2 WSS 3 stored first 4 5

The path of a message

  1. Find a relay. A client asks its endpoint's address which relay host to use, over HTTPS. The libraries send your server's HTTP API calls straight to a relay host too; raw HTTP calls can also go through the endpoint.
  2. Connect. The client opens a WebSocket (wss://) straight to that relay host, which checks its token and the endpoint's limits.
  3. Store. A publish is stored in the endpoint's database and numbered before it is acknowledged.
  4. Read. Echo reads each stored message from the database.
  5. Fan out. Echo sends it to every relay host with subscribers on the channel, and each relay host delivers it to its clients.

Swengle's management for the region starts, watches, scales and replaces the hosts. It is not on the message path: if it is briefly down, the endpoint keeps routing clients and messages keep flowing.

When you publish

When a client reconnects

Scaling within a host budget

A Cluster scales itself on load within its Maximum hosts, which you set (6 by default, up to 22). You are billed for the hosts that run, not for the budget.

A Dedicated server moves to a larger or smaller size, by you or by automatic scaling on its CPU. Shared scales on its own.

What each plan runs on

Shared

Your endpoint on a shared cluster that Swengle runs, with other customers' endpoints and per-endpoint limits. Replicated across zones, with continuous backups.

Shared pricing

Dedicated

One host of your own running the relay, Echo and the database together. A standby host in another zone if you pick Replicated database.

Dedicated pricing

Cluster

A database host, relay hosts and Echo hosts of your own, scaling within your host budget. Up to a complete copy in each of two zones.

Cluster pricing