Hi, If not done so already, It would be useful to run basho bench[0] on the cluster to see what the raw back end performance is
[0]http://docs.basho.com/riak/latest/cookbooks/Benchmarking/ Regards Richard On Thu, Feb 21, 2013 at 2:37 PM, Christian Dahlqvist <[email protected]>wrote: > Hi, > > Does this mean that you are using a single connection to each node in your > test? If this is the case I would recommend increasing the number of > threads as Dmitri Zagidulin recommended in his response in order to get > better throughput. > > Also, what is the size of the objects you are writing? > > Best regards, > > Christian > > > On 21 Feb 2013, at 13:40, catchme <[email protected]> wrote: > > > * What type of hardware are these three nodes? > > Using Redhat UNIX servers > > 4 vCPU - Intel(R) Xeon(R) CPU;31GB memory > > * Can you provide a gist of the script? > > Using Jmeter to send the requests. Have 4 threads sending requests to > these > > nodes continously > > * What client library are you using? > > Java protocol-buffer client > > > > > > > > > > -- > > View this message in context: > http://riak-users.197444.n3.nabble.com/Riak-low-throughput-tp4026917p4026927.html > > Sent from the Riak Users mailing list archive at Nabble.com. > > > > _______________________________________________ > > 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 >
_______________________________________________ riak-users mailing list [email protected] http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
