The technology behind Tornado, FriendFeed's web server - Bret Taylor's blog
http://bret.appspot.com/entry/tornado-web-server

I'm sure IIS is just as good. Right?

Blog post announcing the release of the Tornado web server, based on Python and epoll, providing an asynchronous programming model for web apps. Comes with a web-app framework supporting 3rd-party authentication via Facebook Connect, Twitter, Google, FF, OAuth, and OpenID.

Era un po' che non mi emozionava una tecnologia...

"FriendFeed's web server is a relatively simple, non-blocking web server written in Python. [...] Tornado is an open source version of this web server and some of the tools we use most often at FriendFeed." can handle 8000 reqs/s

interesante

Tornado: Facebook's Real-Time Web Framework for Python

Possible to build a site like FB.. using Tornado Cassandra, Hive (hadoop underneath) memcache, Scribe, Thrift.

Real-time updates have become an important aspect of the social Web that make it easier to share with friends. In March, we introduced a real-time News Feed to make the stream as relevant and engaging as possible for users. Similarly, FriendFeed, which we recently acquired, built their entire site to support real-time updates. It hasn't been easy to build and scale these features, so today we're open-sourcing a core piece of infrastructure called Tornado, which was originally developed by the FriendFeed team. Tornado is a relatively simple, non-blocking Web server framework written in Python, designed to handle thousands of simultaneous connections, making it ideal for real-time Web services. Tornado is a core piece of infrastructure that powers FriendFeed's real-time functionality, which we plan to actively maintain. While Tornado is similar to existing Web-frameworks in Python (Django, Google's webapp, web.py), it focuses on speed and handling large amounts of simultaneous traffic.

starting point for ressources about Tornado python based webserver and tools

This also shows that for massively concurrent problems, such as lots of parallel comet connections, the GIL in Python is a non-issue as we handle the concurrent connections in a single thread. In this post i am going to look at a selection of asynchronous servers implemented in Python.