mountainss Cloud and Datacenter Management Blog

Microsoft SystemCenter blogsite about virtualization on-premises and Cloud

Overview of Cluster Shared Volumes (CSV) in Windows Server 2012 #WS2012 #Hyperv

Leave a comment


Cluster Shared Volumes (CSV) overview:  Windows Server 2012

CSV is a distributed file access solution that provides multiple nodes in the cluster with simultaneous access to the same file system. Virtual machines or applications that run on CSV are no longer bound to storage, and they can share a common disk to reduce the number of LUNs, as shown in the following figure. Live migration of virtual machines becomes faster because volume ownership does not need to change.

CSV does not impose any special hardware, file type, or directory structure restrictions. CSV uses the well-established NTFS file system.

What CSV delivers in Windows Server 2012

In Windows Server 2012, CSV supports new features in addition to the features that are supported in Windows Server 2008 R2.
CSV continues to support virtual machines, and it has been enhanced as follows:

* Expansion of CSV to support more workloads including File Server for scale-out application data, and possibly more

* Enhanced backup and restore of CSV volumes

* Multiple-subnet support

* Simplified set up for CSV by using Failover Cluster Manager

* Support for Server Message Block 3.0 (SMB) file-based storage for Hyper-V

* Better virtual machine creation and copying experience because copying can be performed from any node with the same high performance

* Support for BitLocker volume encryption for CSV volumes

* Improved I/O redirection to support block-level I/O and increased performance that results from multiple-subnet support

* Improved CSV performance by allowing direct I/O for more scenarios

* Support for memory mapped files

* Integration with new file system features such as copy-offload, defragmenting of CSV volumes, and online corruption repair.

CSV proxy file system
CSV volumes now appear as CSVFS. The underlying technology is still the NTFS file system, and volumes are still formatted with NTFS. Because volumes appear as CSVFS, applications can discover that they are running on CSV, which helps improves compatibility. CSV provides a single consistent file name space. Files have the same name and path when viewed from any node in the cluster. CSV volumes are exposed as directories and subdirectories under the ClusterStorage root directory. In Windows Server 2012, CSV uses standard mount points for better interoperability with:

* Performance counters.

* System Center Operations Manager.

* Monitoring free space on CSV volumes.

* File system minifilter drivers (to allow improved interoperability with antivirus software and backup software).

Multiple-subnet support for Cluster Shared Volumes

In Windows Server 2012, CSV has been enhanced to integrate with SMB Multichannel. This helps achieve faster throughput for CSV volumes, especially when they are working in redirected mode. Applications can still have fast data access even with CSV in redirected mode. CSV traffic can now stream across multiple networks, which delivers improved I/O performance when in redirected mode. CSV also leverages SMB Direct (SMB over Remote Direct Memory Access), on network adapters that support this.

Security and encryption support for traditional clustered disks and CSVs

Failover clusters that are running Windows Server 2012 support BitLocker volume encryption for both traditional clustered disks and CSVs. Each node performs decryption by using the computer account for the cluster itself. This account is also called the cluster name object (CNO). This action enables physical security for deployments outside secure data centers and meets compliance requirements for volume-level encryption.

Ease of file backup with CSV

In Windows Server 2012, CSV helps achieve a seamless backup experience for both backup applications and backup requestors. CSV includes:

* Distributed CSV backup infrastructure for Software Snapshot and coordination of CSV snapshot creation.

* Support for parallel backups across CSV volumes and across cluster nodes.

* Application-consistent and crash-consistent Volume Shadow Copy Service (VSS) snapshot support with a full feature set.

* Increased support for both hardware and software backup of CSV volumes.

* Direct I/O mode for snapshot and backup operations.

* Support for using Windows Server Backup (included in Windows Server 2012 RC) for CSV backups.

* Support for incremental backups.

* Full compatibility support for requestors that are running Windows Server 2008 R2.

The following figure shows application-consistent and crash-consistent backup on CSV.

Advertisements

Author: James van den Berg

I'm Microsoft Architect and ICT Specialist and Microsoft MVP System Center Cloud and Datacenter Management

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s