Toad World® Forums

Which benchmark test (TPC-*) is best suitable to see impact on SQL VM vMotion to another host?


#1

Hi,

I am new to BMF test. I like to measure the TPS impact when we vMotion a SQL server VM from one host to another and Database High Availability(HA) test. What is the recommended benchmark test (and configuration) should be used in both cases?

From VMware HA guide, I see some graphs showing the TPS impact during the transition of VM. The TPS measurement is quite flat before and after vMotion occurred (spike occurred during vMotione. See 1,2).

I have tested with TPC-E/H. TPS measurement is never as flat (see 3) as what I saw on the whitepaper. Is something I did not configure properly, or used wrong benchmark test.

Any suggestions is appreciated.

Thanks

http://www.vmware.com/files/pdf/products/vsphere/VMware-HA-Deployment-Guide-Oracle-Database.pdf

(1)

Capture.png

(2)

Capture1.png

(3)

Capture2.png


#2

Not the TPC-H - it simply does 22 queries (all reads). I would do TPC-C with say 5,000 concurrent users - which requires proper commercial BMF license (i.e. 5,000 virtual users).