Comments on: BTC Statistics I

By: (still) nothing clever — Typical Semantic Web Data

(still) nothing clever — Typical Semantic Web Data — Mon, 28 Sep 2009 16:16:00 +0000

[...] my Billion Triple Challenge data-set statistics posts, if you only just got here, catch up on part I, II or [...]

By: gromgull

gromgull — Fri, 18 Sep 2009 16:37:41 +0000

Hi Pramod,

Did you actually read the post? The first line says “using unix command-line tools.” and then I go on about fixing the format. :)

Anyway – I did not use any triple store, I kept the files in n-triples format and I’ve been processing them line by line. A full scan through all triples for some operation takes about 8 hours on my machine. This means I cannot do normal queries, but I can do many other things.

I dont know if any RDF store could load this data, at least not unless you have fantastic hardware (i.e. 64gb ram, etc.), but you may look into 4store, bigdata, and virtuoso that you already have.

By: Pramod

Pramod — Fri, 18 Sep 2009 16:29:09 +0000

Which data store did you use to load the BTC data ?
I am using Virtuoso to load a sample BTC file that is named btc-2009-small.nq
I am unable to load this data as I am having some problems with namespaces, format of the data set etc.

Do you have any suggestions on how to load this data set to a triple store ?

Thanks,
Pramod

By: gromgull

gromgull — Sat, 25 Jul 2009 08:29:44 +0000

No, as I've not mentioned above, it was done using R :)

By: mjmdavis

mjmdavis — Fri, 24 Jul 2009 17:09:23 +0000

That is a beautiful graph… did you use biggles?