Rtree is 100 to,es faster than couchdb

[Stanley Iriele <si...@gmail.com>]

"Rtree is 100 to,es faster than couchdb"

this comment was made today in a meeting when I brought up couchdb's
incremental map/reduce feature especially for time series data.

I thought R-tree was a type of tree and couchdb used some betree or kd tree
for building indexes.

Am i mistaken about this?....I've used couchdb's map/reduce many times and
even when datasets where huge I could still get guaranteed millisecond time
for updated views

Re: Rtree is 100 to,es faster than couchdb

[Stanley Iriele <si...@gmail.com>]

Ahh...phone typo... What he said kinda made me doubt my knowledge... But
I'm glad that what I thought is right
On Oct 1, 2013 9:43 AM, "Jens Alfke" <je...@couchbase.com> wrote:

>
> On Oct 1, 2013, at 7:51 AM, Stanley Iriele <si...@gmail.com> wrote:
>
> > "Rtree is 100 to,es faster than couchdb"
>
> (Is “to,es” a mathematical term or just a typo for “times”? ;-)
>
> I think this comparison is apples vs. oranges, as R-tree is a data
> structure while CouchDB is a program. There is a geospatial index extension
> for CouchDB (https://github.com/couchbase/geocouch/) which likely uses
> R-trees internally; if you need to do lots of fast range searches (in one
> or more dimensions) it would be a good idea to try it out.
>
> —Jens

Re: Rtree is 100 to,es faster than couchdb

[Jens Alfke <je...@couchbase.com>]

On Oct 1, 2013, at 7:51 AM, Stanley Iriele <si...@gmail.com> wrote:

> "Rtree is 100 to,es faster than couchdb"

(Is “to,es” a mathematical term or just a typo for “times”? ;-)

I think this comparison is apples vs. oranges, as R-tree is a data structure while CouchDB is a program. There is a geospatial index extension for CouchDB (https://github.com/couchbase/geocouch/) which likely uses R-trees internally; if you need to do lots of fast range searches (in one or more dimensions) it would be a good idea to try it out.

—Jens