On Tue, 2013-05-07 at 16:02 +0200, Ariel Zerbib wrote:
> But it's only a point of view of an user.

You need to remove the "But" and the "only" from that sentence.

If we are talking about Solr in large-scale deployment (whether we
measure by load, latency requirements, index size or machine counts), it
makes a lot of sense to shun the webapp solutions and treat Solr as
something you install explicitly and something that is prioritized very
high in terms of resources. It could even require root if that would
provide any advantages.

At the other end of the scale, Solr is also used for less intensive
tasks. It might take the role of a secondary functionality for another
webapp (e.g. search for persons in an email application) or the corpus
or user base so small that performance is just not a problem. For these
situations, the "Run it as a WAR together with 5 other WARs"-approach is
sane enough and might be preferable.

What I see in this thread is that there is a lot of focus on the
large-scale view and little on the small-scale. That might be the right
approach, but I really have no clue as to how many Solr installations
falls in either camp.

It is natural that developers sees Solr as the most important component
and that makes us somewhat blind to the situations where Solr is just
another cog in a complex machinery. As the choice of how Solr is
deployed is highly relevant for users and maintenance guys, hearing
their point of view is important.

- Toke Eskildsen, State and University Library, Denmark


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to