This is a guest post by Jordan Quigley-Jones, Product Manager at Pusher.
Pusher is a hosted platform that lets you quickly, easily, and securely build real-time features in your app. You can now add Pusher as an integration on Datadog and monitor your real-time connections and messages.
Over 150,000 developers use Pusher to build features like real-time chat, live UI updates, event broadcasts, or cars moving on a map, by sending messages over websocket connections from a server to clients.
Pusher sends app engagement metrics to Datadog, so you can monitor their activity in real time.
You can use Datadog’s Pusher integration to create unique dashboard views, set up alerts, and monitor Pusher alongside other parts of your infrastructure. Add Pusher as an integration to get started!
Pusher sends 19 metrics to Datadog, including connection rates, message rates, and message sizes. It’s simple to view all metrics aggregated across your account, or you can drill down by Pusher App_ID. Metrics are aggregated and sent every 5 seconds, and will begin populating your history within a few minutes of setting up the integration.
Pusher’s hosted infrastructure makes it easy create a websocket connection between your server and clients to send real-time data using a simple pub/sub system of channels. Developers can use their existing authentication system with private channels for a secure connection.
The Pusher connections metrics on Datadog are broken down by connection type (websocket, SockJS) as well as security (SSL, Non-SSL).
In Pusher, there are several different message types for different purposes, such as Presence events, which are useful for displaying connected users.
The Pusher messages metrics on Datadog are broken down by message type (Broadcast, Client Event, Presence, Webhook, API) as well as size statistics (count, median, average, 95th percentile, maximum).
With Datadog’s ability to correlate and combine metrics in interesting ways, you can now get new insights about your Pusher traffic, such as the messages sent per connection. You can also correlate Pusher metrics with related metrics from other services in your stack to create graphs that were not possible before.
We are excited to roll out this initial set of metrics and look forward to hearing your feedback on Twitter (@pusher) or through support about the integration or new metrics you’d like to see in the future.