[Kernow] xquery on large data: memory increase? or something else?

[Dr. C. D. <du...@te...>]

Hello,

 

I want to run XQuery on medium to large files (from 100 MB to 600 MB of html
code).

 

If I run the query on a portion of these files that is smaller than 100 MB,
everything runs well ... 

As soon as I want to run the query on the complete file, Kernow runs for a
while and then it seems it does not run at all anymore, eventhough I do not
get the hand back.

CPU consumption goes from 30%-50% down to 1%-3%.

Memory consumption went high ...from 700 MB up to 1,3 GB but I have plenty
memory left.

 

I have changed the JRE parameters to -Xmx1024m or even -Xmx2048m  (within
the control panel /java/java/java runtime environment parameter settings)

But it does not change anything ... I still have this magic limit of 100MB
of data I can run an xQuery onto.

 

What can I do? Where can I tell Kernow to effectively use more memory? Or is
the problem another one?

 

I run a windows XP SP3 PC with 4GB of physical mem

 

Thanks

 

Christian