You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@stanbol.apache.org by Dileepa Jayakody <di...@gmail.com> on 2014/11/12 10:22:28 UTC
Clustering Stanbol to improve horizontal scalability
Hi Devs,
When using Apache Stanbol in a production environment, availability and
scalability are important quality attributes.
Can we cluster Apache nodes to scale the service horizontally?
I came across this question on stackoverflow [1] which asks about the
Stanbol scalability and real world applications.
Thanks,
Dileepa
[1]
http://stackoverflow.com/questions/23324663/apache-stanbol-scalability-and-real-world-applications
Re: Clustering Stanbol to improve horizontal scalability
Posted by Dileepa Jayakody <di...@gmail.com>.
Hi Rupert,
Thanks a lot for your insight. :)
(I have missed this email among others hence late reply..)
Regards,
Dileepa
On Mon, Nov 24, 2014 at 4:04 PM, Rupert Westenthaler <
rupert.westenthaler@gmail.com> wrote:
>
> Hi Dileepa,
>
> Sorry for the late response ... but last week I was at the ApacheCon
> and before that I had not managed to answer your question.
>
> Scaling horizontally is possible by duplicating your configuration on
> multiple machines and distributing requests to those.
>
> To share the configuration with different workers the following things
> might be interesting:
>
> * The configuration is provided by the OSGI configuration admin.
> Stanbol uses the Apache Sling installer framework [1] for providing
> configurations. Managing the config in a central JCR Provider could
> work (never tried it).
> * The /datafiles folder can be read-only. So you could provides files
> via a central file system.
> * The Entityhub provides ReferencedSolrServer [2] this can also be
> point to some file path. In that case worker would not need to copy
> stuff provided by the /datafile folder when they start up.
>
> hope this helps
> best
> Rupert
>
>
> [1] http://sling.apache.org/documentation/bundles/osgi-installer.html
> [2]
> https://stanbol.apache.org/docs/trunk/utils/commons-solr#referencedsolrserver
>
> On Wed, Nov 12, 2014 at 10:22 AM, Dileepa Jayakody
> <di...@gmail.com> wrote:
> > Hi Devs,
> >
> > When using Apache Stanbol in a production environment, availability and
> > scalability are important quality attributes.
> >
> > Can we cluster Apache nodes to scale the service horizontally?
> > I came across this question on stackoverflow [1] which asks about the
> > Stanbol scalability and real world applications.
> >
> > Thanks,
> > Dileepa
> >
> > [1]
> >
> http://stackoverflow.com/questions/23324663/apache-stanbol-scalability-and-real-world-applications
>
>
>
> --
> | Rupert Westenthaler rupert.westenthaler@gmail.com
> | Bodenlehenstraße 11 ++43-699-11108907
> | A-5500 Bischofshofen
> | REDLINK.CO
> ..........................................................................
> | http://redlink.co/
>
Re: Clustering Stanbol to improve horizontal scalability
Posted by Rupert Westenthaler <ru...@gmail.com>.
Hi Dileepa,
Sorry for the late response ... but last week I was at the ApacheCon
and before that I had not managed to answer your question.
Scaling horizontally is possible by duplicating your configuration on
multiple machines and distributing requests to those.
To share the configuration with different workers the following things
might be interesting:
* The configuration is provided by the OSGI configuration admin.
Stanbol uses the Apache Sling installer framework [1] for providing
configurations. Managing the config in a central JCR Provider could
work (never tried it).
* The /datafiles folder can be read-only. So you could provides files
via a central file system.
* The Entityhub provides ReferencedSolrServer [2] this can also be
point to some file path. In that case worker would not need to copy
stuff provided by the /datafile folder when they start up.
hope this helps
best
Rupert
[1] http://sling.apache.org/documentation/bundles/osgi-installer.html
[2] https://stanbol.apache.org/docs/trunk/utils/commons-solr#referencedsolrserver
On Wed, Nov 12, 2014 at 10:22 AM, Dileepa Jayakody
<di...@gmail.com> wrote:
> Hi Devs,
>
> When using Apache Stanbol in a production environment, availability and
> scalability are important quality attributes.
>
> Can we cluster Apache nodes to scale the service horizontally?
> I came across this question on stackoverflow [1] which asks about the
> Stanbol scalability and real world applications.
>
> Thanks,
> Dileepa
>
> [1]
> http://stackoverflow.com/questions/23324663/apache-stanbol-scalability-and-real-world-applications
--
| Rupert Westenthaler rupert.westenthaler@gmail.com
| Bodenlehenstraße 11 ++43-699-11108907
| A-5500 Bischofshofen
| REDLINK.CO ..........................................................................
| http://redlink.co/