Your favorite triple store

8 views
Skip to first unread message

Núria Queralt Rosinach

unread,
Dec 1, 2020, 6:30:17 AM12/1/20
to BioHackathon

Dear all,

I would need your recommendation for a triple store with high perfomance and reasoning capabilities. Do you have any favourite? why? license recommendation? Any feedback would be very much appreciated.Thank you so much!

Kind regards,
Núria

Jerven Tjalling Bolleman

unread,
Dec 1, 2020, 7:19:00 AM12/1/20
to biohac...@googlegroups.com, Núria Queralt Rosinach
Hi Núria,

As you don't give any qualifications regarding what is "high
performance".
I would suggest starting with GraphDB (free) by Ontotext.
https://graphdb.ontotext.com/

It has a really nice user interface and quite a lot of in build
reasoning.
Once that no longer suffices you would have enough data to evaluate
alternatives.

Regards,
Jerven
>
> Kind regards,
> Núria
>
> --
> You received this message because you are subscribed to the Google
> Groups "BioHackathon" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to biohackathon...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/biohackathon/b85fd66c-94e0-4acc-a78b-a3dfbf0623f9n%40googlegroups.com
> [1].
>
>
> Links:
> ------
> [1]
> https://groups.google.com/d/msgid/biohackathon/b85fd66c-94e0-4acc-a78b-a3dfbf0623f9n%40googlegroups.com?utm_medium=email&utm_source=footer

--
Jerven Tjalling Bolleman
SIB | Swiss Institute of Bioinformatics
CMU - 1, rue Michel Servet - 1211 Geneva 4
t: +41 22 379 58 85 - f: +41 22 379 58 58
Jerven....@sib.swiss - http://www.sib.swiss

Mikel Egaña Aranguren

unread,
Dec 1, 2020, 8:27:12 AM12/1/20
to biohac...@googlegroups.com, Núria Queralt Rosinach
Hi;

Having worked with Stardog, Virtuoso, and GraphDB, I concur with Jerven, I would go for GraphDB, for the same reasons (Plus it is the cheapest if you are going to go for a high availability cluster)

Cheers



--
Mikel Egaña Aranguren, Ph.D.

https://mikel-egana-aranguren.github.io


Núria Queralt Rosinach

unread,
Dec 1, 2020, 8:29:32 AM12/1/20
to Jerven....@sib.swiss, biohac...@googlegroups.com
Thank you so much Jerven for your fast reply.

Excuse me for the lack of details, by  "high performance" we are mainly interested in query and ETL processes.

n

Missatge de Jerven Tjalling Bolleman <Jerven....@sib.swiss> del dia dt., 1 de des. 2020 a les 13:18:

Jerven Tjalling Bolleman

unread,
Dec 1, 2020, 8:42:00 AM12/1/20
to Núria Queralt Rosinach, biohac...@googlegroups.com
On 2020-12-01 14:29, Núria Queralt Rosinach wrote:
> Thank you so much Jerven for your fast reply.
>
> Excuse me for the lack of details, by "high performance" we are
> mainly interested in query and ETL processes.

That is still a very big lack of details ;) if I where consulting for
you
I would start with these questions.

* how big an ETL.
** How often/what update frequency
** What consistency requirements.
* Types of queries
** how fast must it answer
** allowed deviations in reply time
** service level agreements
** may/can we cache answers
* Budget availability in people hours/hardware/licensing.
* Existing infrastructure available for use.
* Can ETL be avoided by using R2RML etc.
* how much reasoning is
** desired
** wanted
** really needed.
* single store required, or can we federate?

Still, my first answer is a good place to start.

Regards,
Jerven

Núria Queralt Rosinach

unread,
Dec 4, 2020, 11:17:07 AM12/4/20
to biohac...@googlegroups.com
Hi Jerven and Mikel!

Thank you so much for your very useful feedback. I've greatly appreciated it!!!

Take care.
Have a lovely weekend,
n

Missatge de Jerven Tjalling Bolleman <Jerven....@sib.swiss> del dia dt., 1 de des. 2020 a les 14:42:
Reply all
Reply to author
Forward
0 new messages