CDP DC 7.0 (soon 7.1) benchmark/performance testing

[#1]

Hi guys,

I am interested in concepts for benchmark/performance testing of a new CDP 7.x environment.

Does teragen,terasort and others still work with CDP? I would like to stress my HDFS for example.

I was also looking for the "Testing the installation" chapter in CDP DC documentation without success.

It would be great to get some hints from you.

BR André

[#2]

Hi Andre,

For detailed checking disk performance you find here appropriate testing procedures:

https://docs.cloudera.com/documentation/other/reference-architecture/PDF/cloudera_ref_arch_stg_dev_a...

Further good benchmark are here the TCP-DS tests for Impala:

https://github.com/cloudera/impala-tpcds-kit

and for Hive TCP-DS and TCP-H:

https://github.com/hortonworks/hive-testbench

Also recommend to check network performance via Cloudera Manager by the function 

Inspect Cluster Network Performance

Hope this helps,

Regards

Friedel