> In some discussion with local CF developers, we have
> considered what would be the optimal approach for doing
> statistical computing. The consensus seems to be that
> Javascript is too slow, lacks many useful math functions,
> and would in some cases require a lot of data to be
> transferred for client-side processing. ColdFusion offers
> the advantage of server-side processing of data and just
> delivering the results to the user, but might also be rather
> inefficient. A compiled CFX would probably be an improvement,
> but requires the intervention of a whole lower-level
> programming environment. I'm thinking that stored procedures
> might work well in SQL Server, but not everyone thinks so,
> citing lack of many math functions. Some think that
> server-side Java is the way to go. What do you people think?
Statistical analysis, like other data transformations and calculations, is
best performed as close to the data as possible, in my opinion. When I've
had to do things like this in the past, I've used stored procedures to do
this, and the performance has been significantly better than doing it in CF.
I haven't tried server-side Java or CFXs, etc. to do this, but I'd expect
that doing the calculations within the database will give you the best
performance even compared to those options, as they'd all have to export the
bulk of the data, then perform the calculation - within the database the
calculation can be done without transferring unnecessary data to the
database client.
Dave Watts, CTO, Fig Leaf Software
http://www.figleaf.com/
voice: (202) 797-5496
fax: (202) 797-5444
------------------------------------------------------------------------------
Archives: http://www.mail-archive.com/cf-talk@houseoffusion.com/
To Unsubscribe visit
http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk or send a
message to [EMAIL PROTECTED] with 'unsubscribe' in the body.
