On 20/05/13 13:35, Alexander Sicular wrote:
I think the following line is your problem. As others have said, you should not be M/R'ing over an entire bucket the performance will only degrade as you store more items in Riak. You should feed an M/R with results from a search, index or list of bucket/key pairs.
This is something I'm a little confused about as well. I've observed as well that big MR jobs on Riak don't perform well. Yet I thought the idea of MR was that the workload is split up between the nodes, enabling performance to scale as the cluster grows.
On May 15, 2013, at 9:36 AM, kurt campher <[email protected] <mailto:[email protected]>> wrote:/query = riak.RiakMapReduce(client).add('visits2')/_______________________________________________ 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
