Learning HTTP/2

Chapter 3. How and Why We Hack the Web

Using a (relatively) ancient protocol to deliver fast modern web pages has become a virtual act of acrobatics. An entire specialty of web performance engineer has built up around it. One could argue that O’Reilly’s Velocity conference series was born in part from people wanting to share their various tricks and hacks to get the most out of the venerable protocol. To understand where we are going (namely, toward HTTP/2), it is important to understand where we are, the challenges we face, and how we are dealing with them today.

Performance Challenges Today

Delivering a modern web page or web application is far from a trivial affair. With hundreds of objects per page, tens of domains, variability in the networks, and a wide range of device abilities, creating a consistent and fast web experience is a challenge. Understanding the steps involved in web page retrieval and rendering, as well as the challenges faced in those steps, is an important factor in creating something that does not get in the way of users interacting with your site. It also provides you with the insight needed to understand the motivations behind HTTP/2 and will allow you to evaluate the relative merits of its features.

The Anatomy of a Web Page Request

Before we dig in, it is important to have a baseline understanding of what we are trying to optimize; specifically, what goes on in the time between when users click a link in a web browser and the page being displayed on ...