When size of data is large (over 100 MB), Brown-cluster program will be killed. How can I fix this error?

[luongtieumy]

[ajaech]

The size of the input data doesn't matter as much as the size of the vocabulary. How big is the vocabulary you are dealing with?

[luongtieumy]

About 80.000 vocabularies in my data. It's OK now. I think I didn't enough memory for running this program before.
How big is the vocabulary for limited in this program?
Thank you so much!