Graphing Node.js Memory Usage with D3.js and Rickshaw

In our last blog post in our series on using data streaming functionality to create Rickshaw real-time charts, we gave an introduction of Rickshaw, d3.js, and PubNub, and built a basic real-time chart. We strongly recommend checking out that tutorial to learn more about real-time DIY visualizations.

In this blog post (and the video below), we're going to dig a little deeper and build another real-time chart. We'll be monitoring Node.js memory usage using Rickshaw, d3.js, and a handy npm module.

The npm module plugs right into our Rickshaw-PubNub library, and contains everything you need to get started right away. It'll automatically log your memory usage and spawns an express static server.

In the end, you'll have a real-time chart that looks something like this:

How It Works

NodeJS memory usage is exposed through process.memoryUsage().

Every interval_timeout the module publishes the current output of process.memoryUsage() to the suppliedpublish_key and channel on the PubNub network.

Setup

There are a lot of moving parts, but a full working chart example is include out of the box.

First, install the module with npm.

Include in your app with {dev: true}. This will spawn an express server on localhost using the supplied port. This is a simple static server that already includes the contents of pubnub-rickshaw.

That's it! Now when you run your node app, you should see the following:

Visit the url to see a real-time graph of your NodeJS memory profile over time. You can copy this javascript code, modify it, and embed it anywhere. Make sure to set your own channel!

Read more about pubnub-rickshaw-memory in the full documentation.