An administrator receives a complaint that a virtual machine is performing poorly. The user attributes the issue to poor storage performance.
IF the storage array is the bottleneck, which two counters would be higher than normal?
A.
Average device latency per command, in milliseconds (DAVG per cmd)
B.
Average ESXi VMKernel latency per command, in milliseconds (KAVG per cmd)
C.
Average virtual machine operating system latency per command, in milliseconds (GAVG per cmd)
D.
Number of SCSI reservation conflicts per second (CONS per second)
GAVG/cmd – This is the response time as it is perceived by the guest operating system. This number is calculated with the formula: DAVG + KAVG = GAVG
According to this, DAVG will make GAVG too.
Source: http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1008205
Tricky question. Looks like good answers is A and D
http://communities.vmware.com/docs/DOC-9279
section 4.2.2
obsiously DAVG is a good answer.
DAVG – This is the latency seen at the device driver level. It includes the roundtrip time between the HBA and the storage.
DAVG is a good indicator of performance of the backend storage
KAVG has to do with kernel queues, so it’s not relevant.
GAVG = DAVG + KAVG, but we already monitor DAVG, so this won’t give is anything extra as info
Section 4.2.10 Reservation Statistics
CONS/s The number of SCSI reservation conflicts per second.
During the conflict period any I/O going to the LUN will fail with the BUSY status, and will have to be retried. Usually reservations are held for a very short time (few hundred microseconds). Obviously high number of reservation conflicts is not good and I/O latencies will go up when there are retries. Likelihood of having reservation conflicts increases with the number of metadata operations and with the number of ESX hosts sharing the same LUN doing I/O at the same time
Corret – D.