As I have previously mentioned I started the C module for the FAPWS2 web server. That was not to bad however the scope of the C is very small. FAPWS2 goal is to be the fastest WSGI server then there are a lot of things that could be optimized using Cython. Trying to move Python to C would be a very long and slow task with a lot of potential land (memory leak) mines. However to move the Python to Cython would be a much faster and more stable direction after a good amount of profiling.
On a day to day basis I don't think about the early binding performance hit of Python because Python is fast enough. However there are a number of finite things that because of the volume of calls could be a bottlneck for high performance code. I look forward to experimenting with Cython to make FAPWS2 faster.
0 comments:
Post a Comment