«

»

Jan 27 2014

Hands Down the Ultimate VDI/DaaS platform – non-Persistent Desktops – Part 2

Earlier this week in my article Hands Down the Ultimate VDI/DaaS platform – Persistent Desktops I described how Nutanix help organizations to achieve higher performance, maximum capacity and better end-user experience for VDI deployments. This previous article was mostly focused on Persistent Desktops.

I would like now to highlight some of the features implemented in every Nutanix cluster that also benefit non-persistent desktops, but before that I need to go back I time…

clocky

Due to high storage cost adoption barrier for VDI the industry started looking at different ways to reduce capacity related storage costs, and simplify image management. Linked Clones allowed organizations to define a single master image that serve all virtual desktops along with delta individual files that grow overtime. After a defined period of time, which could be anything between hours or months, the desktop was refreshed to its pristine state recovering the storage capacity used by the delta files.

Because Linked Clones were supposed to be refreshed constantly, either resetting to its pristine state or replacing the base image, user profile management became critical for administrators implementing VDI. At the same time the industry started talking about ways to maintain user-installed applications during these refresh operation cycles.

Clearly Linked Clones solved much of the capacity issue, but performance was still a high barrier adoption cost for VDI. The more performance was required the more disk spindles and storage cache was necessary, making traditional storage solutions very expensive.
 
2013-01-21_ssd

With the rise of SSD and Flash technology came the Advanced Tiering. Advanced Tiering allowed administrators to provide a single small tier of SSD to support Linked Clone master images and replica disks, providing much improved read performance with lower latency.

Most recently VDI environments started to be deployed with stand-alone local SSD in each host to support linked-clone desktops that do not demand high availability, eliminating the need for Storage Area Networks for desktop VMs. Many of these implementations also use additional software to maintain user-installed applications during refresh operation cycles.

The truth is that most VDI deployments are a mix of use-cases where both persistent and non-persistent desktops are used concurrently. Additionally, some organizations may prefer to retain the traditional way to manage desktops for both persistent and non-persistent deployments. I also have seen many linked-clone deployments where desktops are rarely refreshed, creating some level of persistency to linked-cloning technology.
 
I won’t revisit the features I highlighted in my previous article but lets re-cap them because they are also remarkable when the subject is non-persistent linked-clone desktops.

  • Scale-Out
  • Heterogeneous Hypervisors
  • Elastic Deduplication
  • Post-process compression
  • Data Locality
  • Native Disaster Recovery and Replication

 
Now that we covered the basics, history and deployment models let’s talk about how Nutanix help organizations with non-persistent VDI deployment.
 

  • Shadow Clones

Shadow Clones intelligently analyze the I/O access pattern at the storage layer to identify files shared in read only mode (ie: Linked Clone Replica).

When a 100% read only disk is discovered, Nutanix will automatically take a snapshot at the storage layer on each Controller VM (CVM) and redirect all read I/O to the local copy.

The below diagram shows what Shadow Copies looks like:

 
The above is a dramatically simpler and more scalable solution than traditional architecture, as the solution will scale indefinitely without degrading performance. Some of the benefits of Nutanix Shadow Copies are:
 

  1. Replica disk data is always served locally to the host (via Extent Cache and Shadow Copies)
  2. Does not require the use of CBRC (Content Based read cache) and is not limited to 2GB RAM (I will discuss this topic in a future article).
  3. Reduced overhead on the Storage Network (IP Network) as read I/O is serviced locally, which ensures the lowest possible latency on the network for both Write I/O & virtual machine traffic.
  4. During boot storms, login storms and antivirus scans all replica data can be served locally and NO read I/O is forced to be served by a single storage controller. This not only improves Read performance but makes more I/O available for Write operations which are generally >=65% in VDI environments
  5. The solution can scale while maintaining linear performance (Performance does not taper off at scale)
  6. When the base image is updated, Nutanix detects the file has been written to an automatically creates a new snapshot which is replicated out to all nodes.
  7. Feature is enabled once and does not require ongoing configuration or maintenance.

 

  • View Composer Array Integration

VCAI is part of the vSphere vStorage APIs for Array Integration (VAAI) stack and allow Horizon View administrators to take advantage of the Nutanix native snapshot feature within the normal administrative workflow of Horizon View and View Composer.
 
Using this feature helps organizations reduce the time taken to provision desktops. When a linked clone pool is created, the cloning operation is offloaded to Nutanix controllers. Nutanix controllers will handle operations such as snapshot creation, clone creation etc. This cuts down the linked clone pool provisioning time and capacity requirements when compared to creating the pools without this feature.
 
Conclusion
These are just some of the architecture features that make Nutanix the best platform for hosting VDI and DaaS. I will slowly start to get deeper into the technology and inner works, as well as highlight the awesome partnerships we are building with technology leaders in the cloud, enterprise, application and management spaces.
 
Thanks to Josh Odgers to allow me to use content from his article, but if you want to learn more visit Josh Odgers blog “Data Locality & Read Cache – Why it’s critical for high performance Horizon View environments (Part 2)
 
Thanks to Josh Odgers for reviewing the accuracy of this article.
 
This article was first published by Andre Leibovici (@andreleibovici) at myvirtualcloud.net.