On 17/5/18 08:21, Jacek Grzebyta wrote:
Hi,
I am trying to compare usefulness 3 rdf triple stores
within Hadoop environment.
I found that:
1. CumulusRDF is dead for now.
2. virtuoso OS is not a player in that place.
3. Jena delivers support for hadoop features. Especially
(reduce (map )) one.
But I have some objections to the local jena:
fragility to bnodes in the dataset, slow complex queries
processing. I do not know if that was solved.
What about rdf4j?
There's no explicit support in the rdf4j core, but there's Halyard
(
https://github.com/Merck/Halyard), which is an RDF4J-compatible
triplestore on top of Hbase.
Rather incredibly they're on release 1.4, and I only just found out
about them a couple of weeks ago. I've been meaning to give it a
spin. If you're going to test it out, I'd appreciate you sharing
your findings here!
Cheers,
Jeen
______________________________