You are the VMware administrator for a community college. You have several virtual machines
that have very read-intensive workloads, and they are beginning to indicate some disk latency
during peak periods. You would like to decrease the storage latency for these VMs. Which of the
following would be a possible solution?
A.
Migrate each of the VMs to a larger datastore
B.
Migrate each of the VMs to a Thin-Provisioned datastore
C.
Configure vFlash for each of the VMs
D.
Delete any snapshots to reduce disk overhead
Explanation:
http://www.joshodgers.com/tag/thin-provisioning/
How exactly does thin provisioning improves disk read speed? I think correct answer is C as vFlash can be used as read cache (if I am not mistaken). Even D sounds more correct than B (Snapshots create multiple vmdk files so it could degrade read speed because you have to jump through different files to read the data).
I agree with you if anything thin provisioning could also slow things down (not eager zeroed). Deleting snapshots could increase throughput. Storage I/O control is probably the best answer although it is not in the list. I would say Vflash or Snapshot deletion are the best answers, certainly not thin provisioning unless they are stating to put each vm on a separate thinly provisioned LUN (on a different set of spindles).
Agree vFlash (answer “c”)
http://www.vmware.com/files/pdf/techpaper/vfrc-perf-vsphere55.pdf
Same, Vflash