You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-user@lucene.apache.org by suresh pendap <su...@gmail.com> on 2017/07/19 05:00:22 UTC

regarding cursorMark feature for deep pagination

Hi,

This question is more about the Implementation detail of the cursorMark
feature.

I was reading about using the cursorMark feature for deep pagination in
Solr mentioned in this blog http://yonik.com/solr/paging-and-deep-paging/

It is not clear to me as to how it is more efficient as compared to the
regular pagination.

The blog says that there is no state maintained on the server side.

If there is no state maintained then where does it get its efficiency from?

Assuming that it does maintain the state on the server side, does the next
page request has to go the same aggregator node which had served the first
page?


Thanks
Suresh

Re: regarding cursorMark feature for deep pagination

Posted by suresh pendap <su...@gmail.com>.
Eric,
Thanks!! for the link.

-suresh

On Wed, Jul 19, 2017 at 8:11 AM, Erick Erickson <er...@gmail.com>
wrote:

> Chris Hostetter has a writeup here that has a good explanation:
>
> https://lucidworks.com/2013/12/12/coming-soon-to-solr-
> efficient-cursor-based-iteration-of-large-result-sets/
>
> Best,
> Erick
>
> On Tue, Jul 18, 2017 at 10:00 PM, suresh pendap <su...@gmail.com>
> wrote:
> > Hi,
> >
> > This question is more about the Implementation detail of the cursorMark
> > feature.
> >
> > I was reading about using the cursorMark feature for deep pagination in
> > Solr mentioned in this blog http://yonik.com/solr/paging-
> and-deep-paging/
> >
> > It is not clear to me as to how it is more efficient as compared to the
> > regular pagination.
> >
> > The blog says that there is no state maintained on the server side.
> >
> > If there is no state maintained then where does it get its efficiency
> from?
> >
> > Assuming that it does maintain the state on the server side, does the
> next
> > page request has to go the same aggregator node which had served the
> first
> > page?
> >
> >
> > Thanks
> > Suresh
>

Re: regarding cursorMark feature for deep pagination

Posted by Erick Erickson <er...@gmail.com>.
Chris Hostetter has a writeup here that has a good explanation:

https://lucidworks.com/2013/12/12/coming-soon-to-solr-efficient-cursor-based-iteration-of-large-result-sets/

Best,
Erick

On Tue, Jul 18, 2017 at 10:00 PM, suresh pendap <su...@gmail.com> wrote:
> Hi,
>
> This question is more about the Implementation detail of the cursorMark
> feature.
>
> I was reading about using the cursorMark feature for deep pagination in
> Solr mentioned in this blog http://yonik.com/solr/paging-and-deep-paging/
>
> It is not clear to me as to how it is more efficient as compared to the
> regular pagination.
>
> The blog says that there is no state maintained on the server side.
>
> If there is no state maintained then where does it get its efficiency from?
>
> Assuming that it does maintain the state on the server side, does the next
> page request has to go the same aggregator node which had served the first
> page?
>
>
> Thanks
> Suresh