You are viewing a plain text version of this content. The canonical link for it is here.

Posted to users@jena.apache.org by Laura Morales <la...@mail.com> on 2017/04/16 13:45:31 UTC

Query performance over 1 dataset, many graphs

If I query my dataset as "SELECT * FROM <some-graph> ...", is there a performance hit if <some-graph> is in the same dataset with many (10s or 100s, more?) other graphs? Or is this fact completely irrelevant?

Re: Query performance over 1 dataset, many graphs

Posted by "A. Soroka" <aj...@virginia.edu>.

To some extent it will depend on the dataset implementation in use, but the two most likely are TDB and TIM (transactional in-memory) and for either of those, there is no particular hit. Both use covering indexes that include orderings to prevent that.

If you are using some other (not core Jena) dataset implementation, it will depend on the specifics. 

---
A. Soroka
The University of Virginia Library

> On Apr 16, 2017, at 9:45 AM, Laura Morales <la...@mail.com> wrote:
> 
> If I query my dataset as "SELECT * FROM <some-graph> ...", is there a performance hit if <some-graph> is in the same dataset with many (10s or 100s, more?) other graphs? Or is this fact completely irrelevant?