Hi Jeremy, I believe that the performance difference for any single query operation (or index operation) would be negligible. I'm making this assumption based on the fact that integers will probably encode into fewer bytes than a binary, but the encoding process is slightly more complex than a binary.
The biggest tradeoff will be in terms of storage space, which would indirectly affect both indexing speed and query speed, especially for clusters with very large numbers of objects. If you (or anybody in the community) is interested in some light Erlang hacking, there is a secondary index driver into basho_bench that currently only supports benchmarking with integers. It should be fairly straightforward to extend this to use binaries, and then do some head-to-head comparisons. The file is basho_bench_driver_2i.erl. Best, Rusty On Wed, Dec 14, 2011 at 11:21 AM, Jeremy Raymond <jeraym...@gmail.com>wrote: > Would an integer based secondary index be faster than a binary secondary > index for exact matching lookups? > > -- > Jeremy > > _______________________________________________ > riak-users mailing list > riak-users@lists.basho.com > http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com > > -- Rusty Klophaus (@rustyio) *Basho Technologies, Inc.* www.basho.com
_______________________________________________ riak-users mailing list riak-users@lists.basho.com http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
