Joshua,

I recently switched from using riak to riak-search in order to index my data
without requiring a map-reduce each time I wanted to query.  For example, my
application needs the ability to filter a set of file objects based on
various attributes such as type and creation time.  Using riak I achieved
this with map-reduce but it required a mapred over a bucket each time which
was just too expensive (I was using the fs backend which didn't help).  To
combat this I created a cache of the results but my application also
requires that a new file is immediately seen by the system which means I was
still running the map-reduce job too often.  I then turned to riak-search
which allowed me to index the data as it comes in (giving the system
immediate visibility of new files) and allowed for very efficient, low
latency queries.

I'd say if you have a job that you are going to run _very_ often for the
purpose of indexing your data then you might want to look at search.

-Ryan

On Sat, Jan 29, 2011 at 10:05 AM, Joshua Partogi <[email protected]>wrote:

> How do you choose whether to use one over the other? What are the
> criterias I should look at?
>
> Thanks heaps
>
> --
> http://twitter.com/jpartogi <http://twitter.com/scrum8>
>
> _______________________________________________
> riak-users mailing list
> [email protected]
> http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
>
_______________________________________________
riak-users mailing list
[email protected]
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com

Reply via email to