=====================================
Django and PyPy: performant is a word
=====================================

Me
---

* Django and PyPy core developer
* I like making **your** code faster
* Working at Quora making their codebase run on PyPy, fast.

What is Django?
---------------

* Anyone here know?

What is PyPy?
-------------

* An implementation of Python 2.7.1
* A very fast implementation
* A very compliant implementation

What is PyPy? (2)
-----------------

* Python written in Python
* Open source (MIT licensed)
* 8 years old
* Over 150,000 lines of test code (that's more than all of Django)
* A successor to Psyco

Fast
----

* Faster than CPython on almost every benchmark we have.
* http://speed.pypy.org/
* A very actively developed project: http://bit.ly/pypy-django-bench

World's shortest introduction to JITing
---------------------------------------

* Run interpreter
* Find frequently executed loops
* Turn those loops into efficient assembler, by specializing for the types
  of variables and other things.

Case studies
------------

* Production ready
* Real people are using this to speed up their apps.

LWN.net
-------

* Parse the output of ``git log`` and generate data/reports
* CPython: 63 seconds
* PyPy: 21 seconds

Some guy on IRC
---------------

* Query PostgreSQL and generate reports.
* CPython: 2 minutes
* PyPy: 8 seconds

Why isn't everyone using PyPy?
------------------------------

* C extensions
* C-API tightly coupled to CPython implementation details

Solutions
---------

* CPyExt
* Pure Python/``ctypes``
* Cython (GSOC)

But web apps are I/O bound...
-----------------------------

* Eh, maybe they should be, but they often aren't.

The Wild Wild Web (WWW for short)
---------------------------------

* To run a Django site you need a handful of things
* Web server
* Database
* Random other libraries (``PIL``, ``lxml``, etc.)

Web server
----------

* WSGI
* Any pure Python server will do
* I like ``gunicorn``, you can use whatever you like
* *Not* ``mod_wsgi``

Database
--------

* Use any database you like, so long as there's an adapter for it that works with both Django and PyPy!

SQLite
------

* Standard library, just works!

PostgreSQL
----------

* RPython ``psycopg2`` compatible lib, requires compiling your own PyPy
* ``pg8000`` and tons of other random libraries, Django doesn't work with them, but if they're pure Python they'll work with other stuff (e.g. SQLAlchemy)

MySQL
-----

* (various expletives censored)
* Nothing that works with Django ATM
* I'm working on a ``ctypes`` based MySQLdb dropin replacement, hopefully open source soonish.

Oracle
------

* We have an RPython ``cx_Oracle``
* I know nothing about its status

Other databases
---------------

* There are other databases?
* Uhh, talk to me later?

Random other libs
-----------------

* ``PIL`` - works under CPyExt
* ``lxml`` - doesn't work :(
* Others - how should I know?  Others isn't very specific.

Benchmarking!
-------------

* Lies, damned lies, and statistics!
* And benchmarks
* Ignore them, you need to test *your* app.
* But if you need to convince your boss...

Django template benchmark
-------------------------

* Part of the Unladen Swallow benchmark suite
* PyPy 1.5: almost 10x faster than CPython
* PyPy trunk: almost 12x faster
* http://bit.ly/pypy-django-bench

Rietveld benchmark
------------------

* Another part of the Unladen Swallow benchmark suite
* PyPy trunk: about 1.35x faster than CPython

Tornado web app
---------------

* 2x as many requests per second

Memory
------

* Mixed bag.
* Some apps use more, some use less.
* Benchmark your own app.

PyPy
----

* A better platform for developing Python itself
* A faster Python for your apps

Recruiting
----------

* We could use some developers/designer to help with our performance tools.
* We have a cool webbased profiling/analyses tool.
* Flask/Jinja/jQuery (sorry)
* Contributors wanted, no compiler experience needed!
* http://bit.ly/pypy-recruiting

Questions?
----------

* http://alexgaynor.net/
* http://pypy.org/
* #pypy on irc.freenode.net
* I want to make your apps faster, come talk to me!
* Thank you!
* Dank je wel!