Nimble Storage and VMware – Virtual Volumes Demo

Nimble Storage and VMware – Virtual Volumes Demo


So, we’ll start off our demo by first registering
the Nimble Vasa provider. We’ll do that by selecting VMWare integration,
and then we have a wizard where we can add VCenter integration. So, we’ve already got one here. We’ll add another VCenter, so we’ll put
in the details for that particular VCenter. And we’re going to select web client because
everything is done through the web clients and also the vasa provider and this will automatically
register the vasa provider with VCenter. And that is successful as you can see at the
top there. Now the next thing we’re going to do is
to create a storage container. We’ll do that in the Nimble UI by creating
a new folder. And for that folder we’re going to select
a management type of VMware virtual volumes and then we’re going to select the VCenter
that we just added integration into. And for this particular folder, we’ll give
it a size limit, and we’ll go ahead and make that five terabytes here. Now we’re going to go ahead and create a
VVol data store. And so, we’ll create new data store of type
VVol. And here we see the VVol demo storage container
that we created earlier, and we’ll go ahead and give it a name. We’ll call it VVol demo. We’ll select the host we’re going to provision
it to, and we’ll click finish. And there we can see our VVol data store has
been successfully created. Click on it here, and we can see the storage
provider for the Nimble array that is backing that. Ok, now what we’d like to do is go and create
some storage polices. And we’re going to use the storage policies
to dictate the behavior of the Nimble array when we create new VVol volumes. So, let’s go ahead and create a new VM storage
policy here, we’ll just leave the default name answer. We’re not actually going to use this, but
we’ll show you the process of creating one. We’ll create rules based on the Nimble storage
data service. This will add a bunch of rules here. First row we’ll add is D duplication, so
with this rule in place all of our VMs will be D duped. Next we’ll add an application policy. And this is something unique to Nimble storage. We’ve got some application policies built
into the Nimble UI, built into the array. So, you can define things like block size,
whether you’re using compression or not. And they’re all custom tuned for specific
applications, and then we’ll just choose that application here. We’ll make sure that it’s an all-flash
array that we’re going to create it on. We can choose to encrypt the data or not encrypt
the data. And we’ve got protection built-in so we
can create policies which will schedule snapshots, which will schedule replication. You can define the number of snapshots you
are going to retain locally, which partner you’re going to replicate to, how frequently
you’re going to replicate, how many snapshots you should retain on that remote replica,
and whether or not you want to, when you delete a snapshot or delete a volume here on the
source side, whether you should also delete that from the downstream replica or not. And we can see that for all those rules we’ve
created on our VVol demo VVol data store is compatible with those, and as we click finish
here our rules will be put into place. Now let’s go ahead and actually create a
VM and use some of the rules that we’ve already created. So, as we give our VM a name here, we’ll
call it VVol demo, and here’s where we get to select our VM storage policy. Now I’m selecting a policy called all-flashed
d duped, which just gives us a generic operating system policy, so we’re going to store the
c-drive on this. For this VM we’re actually going to simulate
what would happen if we created say a Microsoft SQL server VM. So, the C drive, for instance, would have
that base policy of operating system, but we’ve also got two additional drives that
we’re going to add. We’re going to give them some custom policies
here. So, as we can see we’ve got all flash D
duped which is that generic policy we’ve set up, but for the additional drives which
we’re going to add here. As you can see we can have individual policies
for each of those new drives, so we’ll create a new hard drive. We’ll make it 400 gigs, and instead of all
flash D duped let’s use a custom policy for that. We’re going to give it a policy that’s
setup for SQL database volumes. So, this has got the right block size, the
right compression settings, etc. And we’ll do the same thing for a third
hard drive here. We’ll give it a size of 200 gigs and we’re
going to give this one a policy that’s set up specifically for SQL logs. Okay, and we’ll go ahead and hit finish
there, and our virtual machine will be created. There we go. And we’re going to go find that virtual
machine here, we’re going to go ahead and power it on as well. What happens when you power it on is that
swap VVolume gets created so I just want to be able to show you that in the Nimble UI
as well, so we’ll go ahead and power it on here. Alright, flipping back over the Nimble UI
you can see the volumes which are created. We’ve got the VVol demo which is the configuration
VVol which has got all of the meta data, and then we’ve got each of the three data volumes. So, we can see here the first one has that
setting for the base operating system, we’ve got the SQL server data and log volumes here
as well. And then you can see that the V Swap file
there as well. Switching over the protection tab, we can
see that all these share a common volume collection which means they’re going to be a snapshot
in unison and replicated in unison, and we can see those protection settings here. One of the cool things about snapshots with
the VVols is that it’s completely integrated with a storage vendor, so when we take a VM
snapshot directly from the vSphere UI here – what we’re going to end up with is a
native Nimble array snapshot. And if we flip back over to the Nimble array
we can see that that native Nimble snapshot has been taken. We’re no longer dealing with delta files
or anything like that. So, we’ll take a look at the snapshots and
we can see that there’s that snapshot we just created. Cool. So, what happens when we delete a virtual
machine? So, we’ll go ahead and power this one off,
and we’re going to go ahead and delete it from the vSphere perspective. So, we’ll choose delete from disk. So, I just want to show you what happens on
the Nimble array when we do that. So, we’ll go to the Nimble side and we’ll
refresh our view here, and we can see all of those volumes that we once had here are
now offline. They’re not deleted, we’ve still got them
around. We haven’t removed them from the array because
we actually implement a deferred deletion here so we won’t delete those for 24 hours
until after you’ve deleted them from the VMWare side, and that lets you recover those
if you need be. Let’s look at the array here, and we’re
going to just take a look at the CLI. We can get a list of the VMs that it knows
about, and here we can see that VVol demo VM and we can see that it’s been deleted. And in about 23 more hours that’s going
to be automatically reclaimed. Dive a little bit deeper here and see what
else it knows about the VMs. So, we can see all the volumes that have been
associated with it and from here we can either force the destructions so we can reclaim that
space right away, or we’ll go ahead and actually restore this virtual machine using
the VM restore command. So, we’re going to go ahead and restore
this virtual machine. If we pop back over to the web UI and do a
refresh, we can see that those volumes are now recovered, they’re back online and what
we’re able to do know is to scan that VVol data store and add that virtual machine back
to inventory that we just brought back. So, we’ll scan through VVol demo, we’ll
find the VMX file, and we can then add that back to inventory. Now that we’ve added that inventory we should
be able to go back to our host and clusters view and browse for the virtual machine, and
there it is. And that concludes our demo for today. Thank you very much

One Comment

Leave a Reply

Your email address will not be published. Required fields are marked *