@Erick,

I kinda have to learn somehow. And I like to do performance optimizations.
So learning + fun + performance .. yeah.
Other projects that I want to do are harder (our email-exchange, my other
threads on the list) so I have to start on something simpler.

Example:
I have a project where `id` is 3 long fields which are already
indexed+stored (each doc is 10 - 20 long fields (so 1/4 more data for the
id?)), think event-store. The data needs to be in-memory + many
queries/second + records are always increasing etc.





On Sat, Apr 29, 2017 at 9:44 PM, Erick Erickson <[email protected]>
wrote:

> What you haven't really explained is why you care. The only hint I see
> is that you don't want to store the redundant data of your composite
> key as both individual fields and a concatenated <uniqueKey>.
>
> This feels like a whole lot of effort for no particularly good reason.
> Unless you have a bazillion _very_ short docs, concatenating the parts
> together in your <uniqueKey> field is unlikely to be of measurable
> consequence. If you have _measurements_ showing that's not true, it's
> another story.
>
> Assuming I'm on the right path here, I'd guess there are much more
> valuable things you could spend engineering time on than the rabbit
> hole of trying to work around Solr's way of dealing with <uniqueKey>
> ;)..
>
> Best,
> Erick
>
> On Sat, Apr 29, 2017 at 11:38 AM, Dorian Hoxha <[email protected]>
> wrote:
> > @Yonik Thanks, makes sense.
> >
> > @Walter
> >
> > After reading the cwiki page on update request processor looks like you
> > can't modify the GET/DELETE handlers, right ?
> > Because I want to NOT index/store the 'id' field, so clients/solr-cloud
> > (which needs /get) continue to work normally but internally 'id' appears
> > only in translog.
> > And I need to do something similar for GET/DELETE.
> >
> > How to do that ?
> >
> > Regards,
> > Dorian
> >
> > On Sat, Apr 29, 2017 at 8:04 PM, Walter Underwood <[email protected]
> >
> > wrote:
> >>
> >> If you do want a composite key in Solr, you could use an update request
> >> processor script to make it out of the multiple fields.
> >>
> >> wunder
> >> Walter Underwood
> >> [email protected]
> >> http://observer.wunderwood.org/  (my blog)
> >>
> >>
> >> On Apr 29, 2017, at 11:02 AM, Yonik Seeley <[email protected]> wrote:
> >>
> >> On Sat, Apr 29, 2017 at 1:45 PM, Dorian Hoxha <[email protected]>
> >> wrote:
> >>
> >> @Yonik
> >>
> >> Thanks makes sense. So this means that the 'id' need to be indexed(is
> >> always?), (so you can get/update/delete docs not in translog), right ?
> >>
> >>
> >> In Solr, yes.  In Lucene, only if you want lookup-by-id to be fast, or
> >> if you want to use updateDocument with an indexed term for overwriting
> >> documents.
> >>
> >> -Yonik
> >>
> >>
> >> On Sat, Apr 29, 2017 at 7:24 PM, Yonik Seeley <[email protected]>
> wrote:
> >>
> >>
> >> Solr doesn't use Lucene for RT GET, it uses it's transaction log.
> >> Only when the document is not found in the transaction log will it go
> >> and consult the lucene index (which can only search as of the last
> >> commit).
> >>
> >> -Yonik
> >>
> >> On Sat, Apr 29, 2017 at 12:57 PM, Dorian Hoxha <[email protected]>
> >> wrote:
> >>
> >> I know all that. My point is, lucene is NRT, while GET is RT (in both
> >> ES/SOLR). How does lucene return the right document (Term Query) before
> >> doing a commit on GET ?
> >>
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: [email protected]
> >> For additional commands, e-mail: [email protected]
> >>
> >>
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: [email protected]
> >> For additional commands, e-mail: [email protected]
> >>
> >>
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>
>

Reply via email to