Historically, vSphere storage management used a datastore-centric approach. With this approach, storage administrators and vSphere administrators discuss in advance the underlying storage requirements for virtual machines. The storage administrator then sets up LUNs or NFS shares and presents them to ESXi hosts. The vSphere administrator creates datastores based on LUNs or NFS, and uses these datastores as virtual machine storage. Typically, the datastore is the lowest granularity level at which data management occurs from a storage perspective. However, a single datastore contains multiple virtual machines, which might have different requirements. With the traditional approach, it is difficult to meet the requirements of an individual virtual machine.

First we will create a thin provisioned VMFS Datastore that has RAID-5, uses SAS/Flash, FAST Cache and Auto Tiering.

[Click "vCenter Inventory Lists"]

[Click Datastores]

[Click on 'Create a new datastore' icon (left most icon in Objects tab)]

[Click RegionA01]

[Click Next]

[Click Next]

[Click on '<select a host>' dropdown menu.

[Click esx-01a.corp.local]

Determine from the list of LUNs which meets the criteria (Thin Provisioned, RAID-10, FAST Cache, Auto Tiering).
Notice that there is no way to determine this criteria inside vSphere.

[Hover over some of the hosts to highlight accessible disks, then click Cancel]

The Virtual Volumes functionality changes the storage management paradigm from managing space inside datastores to managing abstract storage objects handled by storage arrays. With Virtual Volumes, an individual virtual machine, not the datastore, becomes a unit of storage management, while storage hardware gains complete control over virtual disk content, layout, and management.

View VVol Datastore

[Click on 'Unisphere' tab]

[Click VMware]

Virtual volumes reside inside storage containers that logically represent a pool of physical disks on the storage system. On the vCenter Server and ESXi side, storage containers are presented as Virtual Volumes datastores. A single storage container can export multiple storage capability sets and provide different levels of service to different virtual volumes. Let's review the 'Capability Profiles' to see how this is done.

[Click 'Capability Profiles']

Capability Profiles are used to advertise the available characteristics of a storage pool as part of vSphere Storage Policy Based Management (SPBM). They are configured by the storage administrator and consumed by VI administrators in the form of VM Storage Policies.

Let's look at the Performance Capability Profile by deselecting 'Archive' and selecting 'Performance'.

[Click Archive]

[Click on 'Performance' check box]

Review the 'Performance' characteristics setup by the storage administrator in the right pane.

Take a look at the VVol datastore setup.

[Click on 'vSphere Web Client' tab]

Review VM Storage Policies.

[Hover on the Home icon]

[Click Home]

Let's review a couple of storage policies: 1) VVol-Database and 2) VVol-Logs available for use.

[Click 'VM Storage Policies']

[Click 'VVol Database']

[Click 'Edit Settings']

[Click 'Rule-Set 1']

[Click on scroll bar in 'Drive Type' window']

Review the rest of the Storage Policies.

[Click scroll bar down]

[Click scroll bar down]

[Click 'Cancel']

[Click 'VVol-Logs']

[Click 'Edit Settings']

[Click on 'Rule-Set 1]

[Click the scrollbar down]

[Click Cancel]

At this point even though there is a VVol datastore provisioned that can logically consume every disk on the array, no space has been preallocated until virtual volumes get created.

Create a VM in a VVol datastore.

[Click Home]

Create a VM in a VVol datastore.

[Click 'vCenter Inventory Lists']

[Click 'VM Templates in Folders']

[Click on leftmost icon, 'New VM from This Template']

[Hit any key to type 'sql-01' or click in 'virtual machine' to autofill]

[Hit any key to type 'sql-01' or click in 'virtual machine' to autofill]

[Type 'sql-01']

[Click 'Next' button]

[Click on compute resource 'RegionA01-COMP01']

[Click 'Next']

[Select a 'VM Storage Policy by clicking 'Datastore Default' dropdown menu]

[Select 'VVol-Capacity' from dropdown menu]

[Click 'Next']

[Click checkbox 'Power on virtual machine after creation']

[Click 'Next']

[Click 'Finish']

Review the newly created VM 'SQL-01'.

[Click 'vCenter Inventory L...']

[Click 'Virtual Machines']

Go back to Unisphere and view Virtual Volumes.

[Click on 'Unisphere' tab]

[Click 'Virtual Volumes']

A Config VVol is created to store standard VM-level configuration data such as .vmx files, logs, NVRAM, and so on. One Config VVol is created per VM to store its .vmx configuration file.

A Data VVol is created to store the hard disk. Additional Data VVols are created for additional disks, Snapshots and Clones.

A Swap VVol is created to store a copy of a VM's memory pages when the VM is powered on. Swap VVols are automatically created and deleted when VMs are powered on and off.

Add two additional disks to SQL-01 for DB and Logs.

[Click 'vSphere Web Client']

[Click SQL-01]

[Click 'Edit Settings...']

[Click on 'Select' dropdown menu]

[Click 'New Hard Disk']

[Click 'Add']

Add another hard disk.

[Click 'Add']

[Expand first added hard disk by clicking 'New Hard disk']

[Click 'Vvol-Capacity' dropdown]

[Select 'Vvol-Database']

[Click 'OK']

[Click SQL-01]

In addition to editing the storage virtual machine, the 'VM Policies' can be modified here. Let's change the third disk to the 'VVol-Log' policy.

[Click 'VM Policies']

[Click 'Edit VM Storage Policies']

[Click 'Hard disk 3']

[Click VVol-Capacity]

[Click VVol-Logs]

[Click 'OK']

Go back in 'Unisphere' view the newly created Data VVols.

[Click Unisphere]

[Click the Refresh icon above the list]

Data services, such as snapshots and clones, can be applied at a VM-level granularity by applying them only to the VVols that are related to a specific VM. Also, vSphere Storage APIs for Storage Awareness (VASA) 2.0 allows vSphere to communicate with the storage system to facilitate offloading of these storage-related tasks to the array. Since these tasks are processed by the array, the ESXi host's CPU, memory, and network utilization is reduced. When a snapshot of a VM is initiated using the vSphere Web Client, it is offloaded to the storage system. After the snapshot is created, a new Data VVol is created to store the snapshots contents.
Note: If selected, a memory Snapshot is also created that stores a complete copy of a VM's memory on disk when suspended or for a with-memory snapshot.

Create a Snapshot

[Click 'vSphere Web Client']

Create a Snapshot.

[Click SQL-01]

[Click Snapshots]

[Click 'Take Snapshot...']

[Click the 'Name' field to auto-fill text, or hit any key to type 'Snapshot1']

[Click the 'Name' field to auto-fill text, or hit any key to type 'Snapshot1']

[Hit any key to type 'Snapshot1']

Verify 'Snapshot the virtual machine's memory' is checked.

[Click 'OK']

Verify 'Snapshot the virtual machine's memory' is checked.

[Click 'OK']

View newly created memory Snapshot and additional Data VVols for each snapshotted disk.

[Click the 'Unisphere' browser tab]

Now let's take a look at the VVols snapshot created. Deselect 'SQL-01_2.vmdk' and click on 'SQL-01.vmdk'.

[Click 'SQL-01_2.vmdk']

[Click SQL-01.vmdk]

This is the base VVol the snapshot was made of indicated by the 'Replica Type'. Take a look at the snapshot information of SQL-01.vmdk.

[Click SQL-01.vmdk]

[Click on the first 'SQL-01.vmdk']

Now that we have seen how a VVol is created, let's take a look at deleting a VVol. When a VVol is deleted, there is no administrative work afterwards to free up storage because of remnant metadata.

[Click on vSphere Web Client' tab]

[Click SQL-01]

[Click Power]

[Click 'Power Off']

[Click Yes]

[Click SQL-01]

[Click 'Delete from Disk']

Confirm disk deletion.

[Click 'Yes']

Verify disks have been deleted.

[Click on 'Unisphere' tab]

Space reclamation - Once you delete VVol, it is automatically deleted from the array as well. No further cleanup by the storage administrator.

[End of Demo]