Sebastian, There are plans to have client-driven request routing (that is, at least sending the request to a member of the preflist) in the future, but that is currently vaporware. An interim solution we have discussed was to send the client a "hint" as to where to send a request for that key on the next time around, but that is also unimplemented. On well-performing client and Riak machines, the difference will be small-to-negligible anyway; the primary effect of sending a request to the member of the preflist would be reduced network traffic between the Riak nodes themselves.
On Sat, Sep 8, 2012 at 10:08 AM, Sebastian Cohnen <sebastian.coh...@gmail.com> wrote: > AFAIK Riak does not expose these data via API in order to implement a > "client-driven coordination", right? > > This sounds quite interesting and would be a nice way to reduce latencies > when talking to Riak. > > Best > > Sebastian > > > On 31.08.2012, at 10:55, Dave Brady <dbr...@weborama.com> wrote: > > There's a reference to an article on Basho's site, written by Amazon, about > Dynamo: > > http://features.basho.com/entries/20535121-weighted-nodes > > Section 6.4 explains why they do not use load balancers. > > The rest of the article is good reading, too. > > ________________________________ > From: "Sean Carey" <ca...@basho.com> > To: "Matt Black" <matt.bl...@jbadigital.com> > Cc: "riak-users" <riak-users@lists.basho.com> > Sent: Thursday, August 30, 2012 11:09:00 PM > Subject: Re: Riak behind a Load Balancer > > Matt, > Haproxy is my load balancer of choice. You can always run multiple copies of > haproxy and use some type of dynamic dns with it. > > We do this in many cases. Haproxy scales well. I've seen a single node > sustain multiple gigabits per second with almost no sweat. > > > Thanks. > > > Sean > On Monday, June 25, 2012 at 7:36 AM, Matt Black wrote: > > Dear list, > > Does anyone have an opinion on the concept of putting a Riak cluster behind > a load balancer? > > We wish to be able to automatically add/remove nodes from the cluster, so > adding an extra layer at the front is desirable. We should also benefit for > incoming requests behind shared across all nodes. > > Can anyone see any drawbacks / problems with doing this? > > Thanks > Matt > > _______________________________________________ > 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 > > _______________________________________________ > 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 > -- Sean Cribbs <s...@basho.com> Software Engineer Basho Technologies, Inc. http://basho.com/ _______________________________________________ riak-users mailing list riak-users@lists.basho.com http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
