HTTP's Best-Kept Secret: Caching
Did you know that nearly one-third of the HTTP specification, RFC 2616, is dedicated to caching? Or that the Cache constraint is so fundamental to the REST architectural style that it’s fully defined before “Resources”, “Hypertext”, or “The Uniform Interface” are even mentioned? This talk will show how you can start taking advantage of HTTP’s best-kept secret: caching.
We’ll start by tearing down all of the worst practices that have accumulated over the years and caused caching in HTTP to be perceived as a form of Voodoo. Once deprogrammed, you’ll learn how HTTP’s basic caching mechanisms really work, see how to take advantage of them in Rails, and learn about advanced techniques like “efficient validation”. From there, we’ll talk a bit about the different types of caches (client, shared, gateway) as well as specific cache implementations, like Varnish, Squid, and Rack::Cache. Finally, we’ll show how to think about cacheability as a fundamental aspect of resource design and why it’s beneficial to do so.
People planning to attend this session also want to see:
Ryan Tomayko
GitHub
Ryan Tomayko has been writing about web architecture since 2003 and contributed to Rails core as early as 2005. He created Rack::Cache, is one of the core Sinatra hackers, and is a member of the Rack core team. Ryan works at Heroku in San Francisco helping to build the greatest web platform ever.
Comments on this page are now closed.
Comments
Huge thanks to everyone that attended. I agree that I under delivered a bit and wish I had focused on more advanced stuff. Luckily, the questions at the end were excellent and let us dig in a bit more deeply.
Ryan knows http caching inside and out. Really great talk.
Good clear talk. I agree with the others that it might have been good to go more quickly through the intro, though I know this is often a really tough call for the presenter. The last part of the talk and the Q&A that followed were excellent.
A bit too much on the remedial of HTTP and the caching description could have been covered faster. The bits that described the behavior of request/response headers and the interaction of the gateway cache was quite useful.
Good job. A little heavy on history. Would have appreciated more “best practices” stuff.
Very useful. This stuff should be a 101-level course in our industry… but very few people are aware of it.