[R] question about scalability of R

[Weiwei Shi]

Hi, there:
I am a newbie of R but I am very interested in
applying R in my research. I got some help before from
this mailist and I really appreciate it.

Here is my another question. I think it might be a
common problem for some R's packages.

Recently I used rpart  to do some research and my
dataset has around 200 vars and over 50,000
observations. I loaded half of the data into R (I
cannot load the whole dataset into it, even) and it is
really slow.

I think the implementation of rpart is kinda
memory-resident. I am wondering if there is another
implementation or some way to optimize it to solve the
problem.



Thanks in advance,

Weiwei