Currently Riak is not designed as a multi-tenancy solution. You've hit on the best way, AFAIK, to do this: bucket names as [APPNAME]_[BUCKETNAME]. It will certainly save you hardware costs over having one Riak cluster per application, although I don't know what the performance implications of this might be - MapReduce might get a bit slow if you have a large number of applications with a large number of buckets and a large number keys (100 apps with 10 with 1000 keys is 1,000,000). This problem, admittedly, will go away in Riak 0.14 with the improvements in MapReduce to filter keys by bucket-specific.
Hopefully someone who has built a number of applications on top of Riak can chime in on their real-world experience. Jeremiah Peschka Microsoft SQL Server MVP MCITP: Database Developer, DBA On Wed, Dec 8, 2010 at 11:24 AM, S.A. Kiehn <[email protected]> wrote: > I recently started looking into Riak and like many others I have relational > DB block. My questions are regarding bucket naming and organizing per > application. I have seen examples where the bucket name is something like > "articles". My assumption is that is just an example and that real world > naming would be something that qualifies the bucket some more: > myapp_articles. So if another app that is using Riak has an articles object > then the bucket would be "anudderapp_articles". Is this the way this should > work to separate apps, or am I not understanding something basic. > > Thanks, Scott > > _______________________________________________ > 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
