[pypy-svn] r64911 - pypy/extradoc/talk/euroscipy2009

fijal at codespeak.net fijal at codespeak.net
Fri May 1 04:33:20 CEST 2009


Author: fijal
Date: Fri May  1 04:33:10 2009
New Revision: 64911

Added:
   pypy/extradoc/talk/euroscipy2009/
   pypy/extradoc/talk/euroscipy2009/abstract.txt   (contents, props changed)
Log:
Add an abstract for euroscipy


Added: pypy/extradoc/talk/euroscipy2009/abstract.txt
==============================================================================
--- (empty file)
+++ pypy/extradoc/talk/euroscipy2009/abstract.txt	Fri May  1 04:33:10 2009
@@ -0,0 +1,23 @@
+====================
+PyPy's JIT and Numpy
+====================
+
+Common problem when developing scientific application is that CPython's
+interpreter is too slow. This can mitigated by usage of FORTRAN, C or Cython
+in performance-critical parts of the code, but it would be much nicer
+to just run python faster.
+
+JIT was one of the main objectives of PyPy project for quite a while now.
+As of `March 2009`_ we're able to speed up simple examples in Python
+for about 20-30x over the speed of CPython. Since PyPy's JIT is much more
+flexible than for example Psyco, we expect even greater speedups in the
+future, as well as support for floats and 64bit architecture, which Psyco
+will not achieve in forseeable future.
+
+In this talk we would like to talk about PyPy's JIT, design principles and
+it's future, without going into too many architectural details.
+
+We would also like to present how JIT can make the life easier of simple
+postprocessing of data generated by EULAG model. XXX extend.
+
+.. _`March 2009`: http://morepypy.blogspot.com/2009/03/good-news-everyone.html



More information about the Pypy-commit mailing list