+1 I was thinking of asking the same question.
Concretely, I have been using Riak search to feed data into a machine learning algorithm. For example, we have a query that returns 4M documents and we feed it to the algorithm by iterating through the pages. We've observed a linear decrease in throughput as the page number increases. I noticed today the note about using 2i for deep pagination. I also wanted to try to stream documents out of a map-reduce job with the input set to the search query and an identity function as the mapper. My apologies for hijacking your question Brant! best Marius On 18 August 2015 at 00:20, Brant Fitzsimmons <brant.fitzsimm...@gmail.com> wrote: > Hello all, > > Are the search suggestions on > http://docs.basho.com/riak/latest/dev/using/application-guide/#Search > still valid? > > Specifically, is it still advisable to use 2i when deep pagination is > required, and if the cluster is going to be larger than 8-10 nodes should I > still use something else for search? > > — > Brant Fitzsimmons > > _______________________________________________ > riak-users mailing list > riak-users@lists.basho.com > http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com > > -- --
_______________________________________________ riak-users mailing list riak-users@lists.basho.com http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
