Dear Rajiv, Let me try to address your questions one-by-one:
a) To see the latency/throughput stats and etc., you need to run the workload till the end. Please, see this link if you need more information: https://github.com/brianfrankcooper/YCSB/wiki/Running-a-Workload b) To configure the number of records in the table (during the load phase) you can configure recordcount parameter (e.g., in settings_load.dat). To configure the number of lookup operations (transactions phase), you can change the parameter named operationcount (e.g., in settings.dat). c) I don’t think you can modify the load at runtime. Please, let me know if you have any questions. Best regards, Dmitrii On 04/03/16 11:15, "nishtala" <[email protected]> wrote: >Hi list, > >I did setup YCSB as the client and Cassandra as the server. >With the above setup, I'm checking what is the load-latency-QoS curves, etc. > > >This is how I setup the server: >To start with I did warm up the server using the following command: >java -cp build/ycsb.jar:db/cassandra-0.7/lib/* com.yahoo.ycsb.Client >-load -s -db com.yahoo.ycsb.db.CassandraClient7 -P workloads/workloada >-P settings_load.dat > >and then run the benchmark using this command: >java -cp build/ycsb.jar:db/cassandra-0.7/lib/* com.yahoo.ycsb.Client -t >-s -db com.yahoo.ycsb.db.CassandraClient7 -P workloads/workloada -P >settings.dat -target 10 -threads 10 -p measurementtype=timeseries -p >timeseries.granularity=1000 > >What I fail to find are the following: >a) what is the QoS limit? and how is it defined? >b) How long will the experiment run? I understand it will depend on the >"workloads/ " > can we control that? is there a flag to do that? >c) Is there is a way to modify the load at runtime? > >When you redirect the output from the second command: >it does not show the output for average latency for update and read. > >This is all I get for now: > >YCSB Client 0.1 >Command line: -t -s -db com.yahoo.ycsb.db.CassandraClient7 -P >workloads/workloada -P settings.dat -target 10 -threads 10 -p >measurementtype=timeseries -p timeseries.granularity=1000 > 0 sec: 0 operations; > 10 sec: 100 operations; 9.85 current ops/sec; > 20 sec: 200 operations; 9.99 current ops/sec; > 30 sec: 300 operations; 10 current ops/sec; > 40 sec: 400 operations; 10 current ops/sec; > >is there something I am missing here? > >Any help will be appreciated. > >Best Regards, >Rajiv > > >WARNING / LEGAL TEXT: This message is intended only for the use of the >individual or entity to which it is addressed and may contain >information which is privileged, confidential, proprietary, or exempt >from disclosure under applicable law. If you are not the intended >recipient or the person responsible for delivering the message to the >intended recipient, you are strictly prohibited from disclosing, >distributing, copying, or in any way using this message. If you have >received this communication in error, please notify the sender and >destroy and delete any copies you may have received. > >http://www.bsc.es/disclaimer
