Datastore Contention

Hi Experts,

Hoping you can clear something up for me. What is the easiest way to distinguish between poor disk performance of a disk array, and data contention produced by too many VM's per host per datastore

Would i be right in saying a comparision between datastore latency and the latency of the vms on that datastore would shed such light? Ie if the datastore latency is low by the vm latency is high it would indicate a point of contention?

Kind Regards
LVL 4
Graham HirstIT EngineerAsked:
Who is Participating?
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
Yes, that's the correct approach, check the latency of the VM on the datastore, using esxtop and if greater than 10ms, that VM is having issues with the datastore.

Then change RAID type, (RAID 6 to RAID10) disk type (SATA to SAS, or SAS to SSD)

let me know if you need help with esxtop.
0

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
Graham HirstIT EngineerAuthor Commented:
Fantastic :)

We have tools to monitor vm latency and the datastore latency already, but my main problem was isolating what was contention and what was the disk array not providing enough performance for the requirement.

If i can just look at the datastore, see it is sitting below 10ms but the vm is 20+ then that points me in the right direction of too many vms on that datastore creating a choke point as aposed to fearing the arrays don't have enough iops to keep up with demand
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
Yes, 20ms is higher than the recommended 10ms for a VM.

Baseline the VMs, if it is always this high, if so, you have options above, and moving VM to faster datastore, or change datastore.
0
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
VMware

From novice to tech pro — start learning today.