How can I see performance issues over multiple datastores

[Gabrie1]

Hi

We're getting reports about slow performance during certain hours for a few VMs. These are spread out over a number of datastores, all member of the same storage array. What I would love to have is a view that shows me the sum of IOPs on those datastores instead of each datastore separately. I've already create a custom group that groups all the datastores, but I don't get any performance metrics at that group level only at single datastore level.

Gabrie

[mark_j]

Couple ways to do this.

1. Create a SM on that custom group to aggregate the individ daatstore cmd/sec to a single counter..
2. Graph all of the datastore cmdsec in a metric graph and show a stacked view.. it'll show the total/aggregate as well as the individual breakdown of IOPs.

[Gabrie1]

Hi

I tried to make the super metric like this:

combine(${adaptertype=VMWARE, objecttype=Datastore, attribute=devices|commandsAveraged_average, depth=1})

I now get a "No Data" message in the graph.

When I use sum, I always get a value of 0.

sum(${adaptertype=VMWARE, objecttype=Datastore, attribute=devices|commandsAveraged_average, depth=1})

[mohdhanifk]

VMware KB: Using esxtop to identify storage performance issues for ESX / ESXi (multiple versions)