Chapter 6 - Installation ¶

Up until now, we have used the Blackfire browser extension to profile a project hosted on Blackfire servers. This was the easiest way to get us started, but now it's time for you to use Blackfire on your own projects.

From this chapter on, we will focus on using Blackfire on PHP code. Even though the core message is unchanged, you might need to adapt installation and usage procedures if you develop Python applications.

Infrastructure ¶

Blackfire is a SaaS product. The blackfire.io website allows you to manage your Blackfire project configurations, profiles, and builds. Still, you need to install some software:

The Blackfire PHP C extension, the probe, instruments PHP code and gathers data about runtime behavior. The probe knows how and when to instrument the code, and how to extract data out of the PHP runtime.
The Blackfire agent handles most of the intensive operations (like cleaning, anonymizing, and compressing the data) before sending each profile to Blackfire's servers. It uses a socket to communicate with the probe and HTTPS to send data to Blackfire.

When you are manually triggering Blackfire profiles, you will do so using one of the following tools:

The Blackfire browser extensions make it convenient to generate profiles from a browser (Firefox and Google Chrome are supported).
The Blackfire client, which comes bundled with the agent, lets you generate profiles from the command line. It gives more flexibility and has more features than the browser extensions and will be the covered in a future chapter.

Blackfire servers store some data about your code execution. You can learn about data privacy in our FAQ and in this blog post.

Installation on your Local Computer ¶

Blackfire is supported on many platforms; install it by following the interactive instructions for your operating system:

Installation on Staging, Testing, and Production ¶

This tutorial can followed along whether you have an active subscription or not, as all applications used in this book are publicly profilable.

You may want to go the next chapter if you don't currently have a subscription.

When deploying code to testing, staging, or production environments, you can use the same installation procedure as in the previous section, or use Blackfire's integrations with Chef, Puppet, Ansible, or Docker.

Blackfire is also pre-installed on some platforms, such as Adobe Commerce Cloud or Platform.sh. The configuration procedure might be a little different, please make sure to refer to those platform's documentation.

Next Steps ¶

You are now ready to profile your very own projects!

First, validate your installation by generating a profile from a browser extension (Firefox and Google Chrome). If you have any problems, read our troubleshooting guide or contact our support.

Once you're able to generate a profile, use the profiling methodology we described in the previous chapters to run Blackfire on your applications. I'm very confident that for any non-trivial codebase, you will find some optimizations.

Found an optimization? Share it with us on Twitter and use the #blackfireio hashtag.