Excerpts from chung's message of Mon Oct 04 17:12:18 +0200 2010:
> Hi Christoph,
>    What is dimension (number of features and compounds) in this dataset?

If I remeber correctly (both datasets have been deleted in the meantime)
dataset 3 had ~ 90 compunds/1 feature, dataset 5 ~ 5000 compounds/1
feature.

I have the impression that our processing times are roughly similar. 
In our system things started to get nasty (1) with a larger number of
features (> 1000), (2) with complex features using Tuples. I will try to
generate an example dataset for you to try (might take a little bit
...).

Would it be hard for you to switch to "plain" RDF (i.e.  omit all
RDF:type statements) and compare computation times with OWL
representations.

Best regards,
Christoph