1) Common sense approach to BCP
2) Nutanix DR – Replication and SRM like features
3) 3rd Party Replication – Veeam & Zerto

Don’t let this be you. Click on the picture to register.

Before starting at Nutanix I wrote a brief article on Medusa, Nutanix: Medusa and No Master. Medusa is a Nutanix abstraction layer that sits in front of a NoSQL database that holds the metadata of all data in the cluster. The database is distributed across all nodes in the cluster, using a modified form of Apache Cassandra. As virtual machines move around the nodes(servers) in the cluster they know where all their data is sitting. The ability to quickly know where all the data is sitting is why hard drive failures, node failures and even whole blocks* can fail and the cluster can carry on.

When a file reaches 512K in size, the cluster creates a vDisk to hold the data. Files small than 512K will be stored inside of Cassandra. Cassandra runs on all nodes of the cluster. These nodes communicate with each other once a second, using the Gossip protocol, ensuring that the state of the database is current on all nodes.

A vDisk is a subset of available storage within a container. The cluster automatically creates and manages vDisks within an NFS container. A general rule is that you will see a vDisk for every vmdk since most times they are over 512K. While the vDisk is abstracted away from the virtualization admin it’s important to understand. vDisk’s are how Nutanix is able to present vast amounts of storage to virtual machines with only having a subset of the total amount on anyone node.

vDisks are Block-level devices for VMDKs,
These are mapped seamlessly through the Nutanix NFS Datastore

vDisks are made up of extents and extents groups which help to serialize the data to disk. This process also helps avoid misalignment issues with older operating systems. All of the blocks that make up a vDisk are maintained my Medusa. As workloads migrate between flash and HDD automatically, consistency is maintained across the cluster. If hot data is in flash on one node in the cluster, it’s replica is also in flash on another node, vice versa if the data is stored on HDD.

Cassandra does depend on Zeus to gather information about the cluster configuration.

* You need three blocks before you can survive a whole block going down, we call this feature block awareness.

For your GPU users you want to match the MaxAppFrameRate with the PCoIP frame rate. If your GPU is rendering a workload higher than the PCoIP Max you’re wasting resources for no real gain.

MaxAppFrameRate
HKLM\SOFTWARE\VMware, Inc.\VMware SVGA DevTap\MaxAppFrameRate

PCoIP
HKEYLOCAL_MACHINE\SOFTWARE\Policies\Teradici\PC0IP\pcoip_admin\pcoip.maximum_frame_rate

For more performance information read this whitepaper from VMware,
http://www.vmware.com/files/pdf/view/vmware-horizon-view-hardware-accelerated-3Dgraphics-performance-study.pdf

Zookeeper runs only on three nodes on the cluster, no matter how big or small the cluster gets. Since it’s tracking configuration data that doesn’t change the often there is no impact on performance. Using multiple nodes prevents stale data from being returned to other components, while having an odd number provides a method for breaking ties if two nodes have different information. One Zookeeper node is elected as the leader. The leader receives all requests for information and confers with the two follower nodes. If the leader stops responding, a new leader
is elected automatically. You can easily tell who the cluster leader is by doing a cluster status. Nutanix uses Paxos-like algorithm for consistency.

The cluster leader will also be the node that will be the point man for the support team. The support team comprised of Rock Stars from VMware, Oracle, Cisco and on and on seat right beside engineering for fast response. The cluster leader’s entry point into support can be shut off for cold sites but does provide a available link to support.

Zookeeper has no dependencies, meaning that it can start without any other cluster components running. More info on Zookeeper

Part 3 will be our Distributed Metadata layer with Medusa & Cassandra

Below are core services and components that make NDFS tick. There are actually over 13 services, for example our replication is distributed across all the nodes to provide speed and low impact on the system. The replication service is called Cerebro which we will get to in this series.

This isn’t some home grown science experiment, the engineers that wrote the code come from Google, Facebook, Yahoo where this components where invented. It’s important to realize that all components are replaceable or future proofed if you will. The services\libraries provide the API’s so as newest innovations happen in the community, Nutanix is positioned to take advantage.

All the services mentioned above run on multiple nodes in cluster a master-less fashion to provide availability. The nodes talk over 10 GbE and are able to scale in a linear fashion. There is no performance degradation as you add nodes. Other vendors have to use InfiniBand because they don’t share the metadata cross all of the nodes. Those vendors end up putting a full copy of the metadata on each node, this eventually will cause them to hit a performance cliff and the scaling stops. Each Nutanix node acts a storage controller allowing you to do things like have a datastore of 10,000 VM’s without any performance impact.

While the diagram can look a little daunting, rest assured the complexity has been abstracted away for the end user. It’s a radical shift in data center architecture and will be fun breaking it down.

At a feature level, Direct Desktop Launch (DDL) is my prized gem for this release. From a Nutanix perspective I can take 3 or 4 nodes in a separate cluster and quickly figure out the impact of hardware or software changes on my VDI environment with out a lot of required infrastructure. Take the number of desktops I can run after the change and divide by the number of nodes I was testing with. So even if I have 5,000 VDI production environment, I really only need to maintain 4 node test environment to get a deep understanding of a pending change on my environment. I think customers should be looking at this tool moving forward.

Login VSI v4 is almost a complete re-write of Login VSI 3.x. For Login VSI v4 four major design goals were set:
o Dramatically reduce and streamline installation footprint of VSI. Login VSI v4 is now much easier to install and integrate: in both lab and production environments.
o The test image footprint of Login VSI has been reduced by almost 90%. This makes the tool not only easier to install, but also easier to integrate and deploy.
o Centralization of management, updates and logging makes the use of Login VSI more efficient than ever.
o Direct Desktop Launch (DDL) mode enables large-scale testing with minimal infrastructure.

Further simplify test configuration and customization: benchmarking and load testing of VDI and SBC environments was never so easy and accessible.
o The test image footprint of Login VSI has been reduced by almost 90%. This makes the tool not only easier to install, but also easier to integrate and deploy.
o Centralization of management, updates and logging makes the use of Login VSI more efficient than ever.
o Direct Desktop Launch (DDL) mode enables large-scale testing with minimal infrastructure.
o Substantially increase test realism: no ‘synthetic’ VDI and SBC performance benchmark was ever so ‘realistic’.
o The duration of the standard workloads has been increased from 14 to 48 minutes loops. Also the way in which segments and applications start has been improved to better reflect real world user behavior.
o The datasets used in the workloads now offer 1000 different documents per type, more and larger websites, and a video library in every format, all to ensure a real world variety in data usage.
o The execution of the workloads is improved through the introduction of phasing, allowing for real world production user scenarios.

Workload related changes
 Completely revised and up-to-date standard Login VSI workloads
o More realistic user workload patterns
o All workloads are now a 48 minutes loop (instead of 14 minutes)
o Each workload consists of 4 x 12 minute segments
o Alternating initial segment for each session to ensure equal load distribution
o HTML5 video player instead of flash (flash player is still optional)
 Central data server for content in workload (instead of data in Program Files)
o More realistic data set and data/file access in Login VSI workload
o Larger document files in bigger Pro data pool: 1000 documents per type (Login VSI
Express contains 50 files per document type)
o More and larger websites: each website contains 20 variants
o Websites can run directly from share or through a IIS server
o MP4 video library with all formats: 480p, 720p and 1080p
o Login VSI v4 Pro library now +12GB (3.7: 300MB)
o Multiple content servers can be configured for large scale tests
Centrally manage and start launchers (zero-touch)
o No installation of Login VSI launcher software required
o Automatic launcher start with RDP desktop sessions
Fully customizable and exportable AD setup in PowerShell script format

New Direct Desktop Launch (DDL) Mode: start workloads without a remoting client required
[Pro only]
o Minimum launching infrastructure required for testing
o Start thousands of desktop sessions from a single launcher
o Supported on Windows 7 and Windows 8
o Useful for comparing & Storage configurations, Server or hypervisor changes

* Both offer top class Time to Value. With Nutanix’s ability to go from shrink wrapped to production in 1 hr or less paired with easy to setup replication from Veeam. Remote sites can be easily sprung up. Set and forget.

* IT teams don’t have double the time to setup DR facilities. Veeam gives the ability to provide replication from legacy storage arrays to the Nutanix Complete Cluster

* Veeam with use of Backup Proxies can distribute replication jobs on a Nutanix cluster so continues replication can be achieved with lowest impact to a running workload. See this thread for possible bad outcomes otherwise -> http://forums.veeam.com/viewtopic.php?f=2&t=10764 (Vendor does have VAAI now, but the risk remains)

* Nutanix can offer the greatest about of capacity and performance for Veeam Instant Recovery. DR volumes on Nutanix can be set to receive more IOPS for ensuring recovery.

* Nutanix can offer global compression across to cut storage costs.

* Nutanix’s distributed flash allows recent changes to stored in flash for fast backup.

I’ve made serious efforts with VMworld sessions in the past with out much luck. I am very hopeful this year though. My session this year – 5060 Benchmarking the Horizon Workspace Appliance with Performance and Scalability – is pretty cool. The session is a chance to showcase new a approach to load testing the Horizon Workspace. I hope I get the chance to co-present with Manrat Chobchuen from VMware.

If doesn’t happen it will be great for my local VMUG!

Pick want you want to see at VMworld while the voting is open

“By 2017, the major public cloud compute architectures will be common architectures in enterprises.”

“The greatest transformation from the cloud will come from true scale-out application architectures.”

Gartner Data Center Conference, Keynote: Rethink Infrastructure and Operations to Dramatically Reduce Costs, Raymond Paquet, December 2012.

Nutanix is helping to bring Private Cloud to the Enterprise by embracing the Software Defined Data Center(SDDC) story. The engineers of Nutanix, which consist of people from Google, Aster Data, Facebook, Oracle, Yahoo(should be enough name dropping) have helped to bring all the intelligence out of big iron hardware and into software so it can be quickly scaled up and out. In essence they have flattened the data center, combining storage and compute in one easy nice pill to sallow. The pill happens to be in dense block form factor but the magic is in the software.

The magic allows for:

* Highly scalable compute and storage resources

* Able to scale up and down, scale up is the hardest problem to solve and scale down allows you to shift physical resources with two commands

* It just works. No drivers needed in VM’s to enjoy our “Hot Cache”. Hot cache is vMotion friendly, no re-hydration needed.

* Giant Data store with no VM limits. This is the Cloud slayer! Adding compute and storage and not having to go into the application stack to configure anything. Application folks can quickly get resources added for their projects. Being quick on the draw stops folks from pulling out their credit cards for Amazon.

* Just like the best Public Clouds, you pay as you grow. Start small, new projects comes online just add another node. Boom, Done.

* Multi-hypervisor support

* The intelligence of the systems lives beside the VM’s. Able to add extra cache with a click of the mouse.

* When new features are added, all platforms get to enjoy it since there is no dependence on hardware