Need an SVM implementation or a Java library -

I have a data set with 2400 samples and 10,000 features. All data is binary (+1 or -1). I need to run it with an SMM algorithm, so I can compare it to my algorithm. However, I do not have much information about SVM or which package to use, I tried to read about it so I could apply one, but this is a way to my head, and I received from him Weight vector I am a Windows user and I have got my implementation in Java. I could export my data in a text file with 1 to 0 and 0 I have access to matlab, but some tell me that it will be very slow and in fact my 1.6 hour 2 GB RAM laptop And I need it) will not be fast enough. I need to run the algorithm two hundred times to get the exact result.

I am really looking for a quick and easy-to-understand library or SMM implementation which I can use in my case.

Thank you all. Feel free to ask any additional questions to help me better.

I ended up using JNI for SVN light which can be found here:

How to use it, it did not take long and it is amazingly fast (seconds).