On Wed, Dec 15, 2010 at 2:02 PM, Jonathan Hartley <tart...@tartley.com> wrote: > Can I ask you to clarify one thing you mentioned? My understanding was that > speed of numerical modelling was only of such vital import if you are doing > low-latency automated trading, in the sort of scenario where you need to be > on a box placed on a LAN in proximity to the trading server, in order to > make sub-millisecond trades. On the other hand, if you're running across the > internet, then any slowdown due to using Python verses another language > would be vastly swamped by network and other IO delays. Am I very much > mistaken?
You're right that a lot of financial calculations are used in areas outside of low-latency, but even for end of day batch procedures these things need to be optimized purely because we're talking about huge amounts of computation here. Calculations that can take hours even in fortran or c. Plus due to the large amount of legacy libraries in Fortran/C++ it tends to mean those languages are favoured even when speed isn't critical. Most Python usage I've seen in finance has been as a glue language often replacing Perl or Bash scripts or to do rapid prototyping, especially in conjunction with tools like QuickFix. Katie -- CoderStack http://www.coderstack.co.uk/python-jobs The Software Developer Job Board _______________________________________________ python-uk mailing list python-uk@python.org http://mail.python.org/mailman/listinfo/python-uk