you could always use a benchmark like YCSB https://github.com/brianfrankcooper/YCSB/wiki with a write workload to get a clearer view of you systems throughtput. You can configure the size of the data being inserted (1500 bytes if you are talking about IP packets) to match to your needs.
Regargs, Panagiotis Garefalakis On Mon, Dec 23, 2013 at 10:05 PM, Aaron Morton <aa...@thelastpickle.com>wrote: > nodetool proxyhistograms shows the throughput for the node, nodetool > cfhistograms shows it for a single node. > > If you want to get an overview install something like Ops Centre > http://www.datastax.com/what-we-offer/products-services/datastax-opscenter > > Cheers > > ----------------- > Aaron Morton > New Zealand > @aaronmorton > > Co-Founder & Principal Consultant > Apache Cassandra Consulting > http://www.thelastpickle.com > > On 19/12/2013, at 8:46 pm, Jason Wee <peich...@gmail.com> wrote: > > Hello, you could also probably do it in your application? Just sample with > an interval of time and that should give some indication of throughput. > > HTH > > /Jason > > > On Thu, Dec 19, 2013 at 12:11 AM, Krishna Chaitanya < > bnsk1990r...@gmail.com> wrote: > >> Hello, >> >> Could you please suggest to me the best way to measure write-throughput >> in Cassandra. I basically have an application that stores network packets >> to a Cassandra cluster. Which is the best way to measure write performance, >> especially write-throughput, in terms of number of packets stored into >> Cassandra per second or something similar to this??? Can I measure this >> using nodetool? >> >> Thanks. >> >> -- >> Regards, >> BNSK >> *. * >> > > >
