bulk loading into named graph

2 views
Skip to first unread message

jesse....@wur.nl

unread,
Jul 24, 2014, 8:43:34 AM7/24/14
to sta...@clarkparsia.com
Hi,

I would like to bulk load a dataset into a named graph, but I can not find the option to do so.

How much faster is the bulk load actually in comparison to the add data function?

Thanks,
Jesse van Dam

dfwin...@gmail.com

unread,
Jul 24, 2014, 9:04:57 AM7/24/14
to sta...@clarkparsia.com, jesse....@wur.nl
We've used TRIG formatted files to bulk load to named graphs during db create, seems to work fine and is quite a bit quicker than add data.
Dave Winchell

Mike Grove

unread,
Jul 24, 2014, 9:09:24 AM7/24/14
to stardog, jesse....@wur.nl
On Thu, Jul 24, 2014 at 9:04 AM, <dfwin...@gmail.com> wrote:
We've used TRIG formatted files to bulk load to named graphs during db create, seems to work fine and is quite a bit quicker than add data.
Dave Winchell

As Dave said, you can load any RDF format which supports named graphs natively, such as TRiG, directly from the command line.  The bulk loader does support the ability to add non-named graph supporting formats, such as Turtle, into a specific named graph, but that functionality is currently only available via the Java API [1].

Bulk loading is significantly faster than `data add` because the latter has to allow for reads which may take place within the transaction, because of that overhead, you will see much better performance adding data when the database is created rather than transactionally to an existing database.  We're working on changes to create a write-only transaction whose performance will be close to our bulk load speeds.

Cheers,

Mike

 


On Thursday, July 24, 2014 8:43:34 AM UTC-4, jesse....@wur.nl wrote:
Hi,

I would like to bulk load a dataset into a named graph, but I can not find the option to do so.

How much faster is the bulk load actually in comparison to the add data function?

Thanks,
Jesse van Dam

--
-- --
You received this message because you are subscribed to the C&P "Stardog" group.
To post to this group, send email to sta...@clarkparsia.com
To unsubscribe from this group, send email to
stardog+u...@clarkparsia.com
For more options, visit this group at
http://groups.google.com/a/clarkparsia.com/group/stardog?hl=en

Reply all
Reply to author
Forward
0 new messages