Matlab Real Time Stock Data Feed Efficiency

We are interested in using Matlab to do Real Time Stock Analysis and Algorithm Processing. This application would be inhouse on the LAN. We have Windows 2008 and SQl servers to receive the data. If the data is real time we obviously want to get it into Matlab as soon as possible for algorithm and or pattern analysis.

What method would be fastest for integration to get realtime performance?  Would the data be stored in text/M files or database then accessed or could it direct connect to RT data for dynamic processing.?

Please provide links to strategy documents or videos?
LVL 7
XGISAsked:
Who is Participating?
 
TommySzalapskiConnect With a Mentor Commented:
If the data is dropped into the database real time, then just connect query the database directly from MATLAB.
Details here: http://www.mathworks.com/help/toolbox/database/ug/database.html

If you want speed though, MATLAB might not be the best tool. It's a lot slower than C/C++/C# (or basically any compiled language). C# integrates well with SQL Server and is easy to learn although you'd probably need a math library.

At least remember to declare the sizes of all your arrays before building them. Dynamically growing arrays is the most common mistake in MATLAB that slows it down.
0
 
XGISAuthor Commented:
Gday Tommy. Thanks for that link. We are primarily .NET C# programmers and maintain our data in SQL 2008 R2 so that is a good thing.
Cheers Aaron
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.