HP 3PAR StoreServ Cluster Extension Software Administrator Guide (5697-2442, July 2013)

Cluster Shared Volume for Windows Server 2012
Cluster Shared Volume is a feature of Microsoft Failover Cluster which allows all nodes in the
cluster with the ability to directly access the same volume without changing ownership of the disk
resource. The result of the feature is that all nodes in a cluster can use the same volume to host
actively running Virtual Machines at the same time. CSV manages storage access differently than
regular clustered disks. CSV Volume is a shared disk containing NTFS partitions. CSV gives you
the ability to store multiple VHDs on a single LUN and run the associated VMs on any cluster node.
This will make your cluster work quicker. Additionally, CSV enables Live Migration which allows
you to move a running VM from one node to another node with zero downtime. Since disk
ownership no longer needs to change when a VM moves to another node, this makes the process
quicker and safer, allowing clients to remain connected while the virtual machine is moved to
another node in the cluster. The CSV disk resource which is owned by a cluster node is called as
coordinate node. The VMs which are hosted on the CSV disk can be owned by either coordinated
node or any other node of the cluster. When a disk is in the available storage group, it can be
converted to CSV. When a CSV disk is created in the Failover Cluster, internally, a hidden role
will be created by the failover cluster which is not visible in the Failover Cluster GUI, but visible
through cluster.exe command output. Only CSV disk is shown in the Failover Cluster GUI. The
name of the CSV role is in the form of a GUID. For every highly available VM created on the CSV,
there will be a role created in the Failover Cluster which is independent of the CSV role. The
following screenshot has two virtual machines created on a single CSV Disk. For each virtual
machine there is a role created.
In a multi-site storage array replicated environment, CSV disks are replicated to the remote
datacenter using the array based remote replication feature. VMs residing in the CSV disk can run
on any cluster nodes (physical servers). In case of disaster to the primary datacenter, CSV disks
need to be brought online in the secondary datacenter. For this, storage failover needs to be done
after checking the CLX disaster recovery rules before cluster brings the CSV disk online. This is
where CLX comes into picture to swap replication direction and perform the storage failover in an
automated fashion to add failover capabilities for the CSV disks in the clustering environment.
Planning for HP 3PAR StoreServ Cluster Extension 15