This post has already been read 21340 times!
Nutanix’s mission is to virtualize a datacenter without the need for a SAN. This approach allows for many benefits, such as buy what you need, when you need it and a reduction in complexity around architecture and operations.
The Nutanix NX3000 comes per block of 4 blade like servers called ‘nodes’. Each Node has local storage comprising of 1x PCI-e SSD (400GB) , 1x 2,5” SATA SSD (300GB) and 5x 2,5” SATA HDD (1TB). All the nodes have two CPU sockets. Each socket can now contain 8 core CPUs. This gives a total of 16 physical cores per node and 64 physical cores per block. Each node is configured with 128Gb to 256Gb Memory . With a maximum of total configured memory of 1TB per block. The NX-3000 series is now configured with dual 10GbE cards.
With this one block configuration Nutanix will now support up to 400 VMs. It replaces 10U equipment with a single 2U block.
Nutanix NX-3000 now offers a new VM-aware Disaster-Recovery solution which allows VMs to be grouped together into what are termed protection groups. These groups of VMs can then be snapshotted and replicated to a Nutanix block at a remote site.
Nutanix OS 3.0 also delivers runbook (failover and failback) automation that is hypervisor-agnostic, which means the native disaster recovery capabilities are available and consistent regardless of the underlying virtualization platform or management tools.
The runbook policies are flexible, and could include doing master-to-master replication and not just master-to-slave. So every site can act as both a primary and a backup site. This provides a lot of flexibility in setting up disaster recovery policies.
Nutanix supports both in-line and post-processed compression. In-line (online) processing is good for sequential processes like archival or hadoop. This compression technique works on writes, reducing the amount of data which needs to be stored, but is very CPU intensive.
Post-processed (offline) compression has less impact on CPU and SSD performance. This is ideally suited for random, batch workloads, and it avoids impacting the current workload/IO path. This uses map reduce techniques to essentially figure out which data has gone cold and then compress it. This technique is based on Google’s snappy compression library and hits a nice balance between high-speed & reasonable compression rate
In-line compression is very common, Post-processing compression is unique to Nutanix.
The software, which was designed to be hypervisor agnostic, now supports KVM and VMware vSphere 5.1 where the previous version only worked in VMware environments.
Other Hypervisors including Microsoft Hyper-V and Citrix XenServer will be supported in the future.
Nutanix OS 3.0
NOS 3.0 lets customers upgrade storage nodes dynamically without any downtime, with workloads on the node to be upgraded dynamically, moving to other nodes while the upgrading is being done. Nutanix OS (NOS) 3.0 is now based on Centos.
Nutanix Controller VM
iSCSI / NFS
Nutanix was built around iSCSI but now have the ability to run NFS as well. Nutanix calls their implementation of NFS, NDFS. With their release of NDFS they have re-written the standard NFS code to turn allow for better performance. NFS was built to house thousands and thousands of files, today with virtualization it’s almost the opposite. With VMware we have large files, known as VMDK’s that containerize all the smaller files. Nutanix has tuned their implementation of NFS to work with few files large files instead of the old paradigm. NDFS can localize NFS client-server communication giving it the shortest path on the IO highway at all times.
VMware vSphere Integration
Full support for vSphere 5.1!. Nutanix also supports VAAI (vStorage APIs for Array Integration ) and VCAI (View Composer Array Integration). Nutanix also provides their own command line interface called nCLI.
The NX-3000 is priced at $144,000, which includes four server nodes in a single enclosure. The disaster recovery and the new post-process compression require a separate software license.
Storage Field Day 2 video’s
Below are 2 very helpful video’s where Mahit Aron and Steve Poitras explain the complete Nutanix stack.
After 15 minutes talking about the Nutanix product as it exists today, Aron dives into new features: Native disaster discovery and data compression.
Nutanix demo: deploying a new from factory Nutanix cluster, configuring storage and exposing it to ESXi hosts, provisioning a 400 desktop view pool while running a Hadoop job, scaling out the cluster from 4 to 8 nodes.
Nutanix, the start-up that came out of stealth mode in 2011 really build a name for themselves last year. In my opinion they uncomplicate your datacenter, you no longer need a seperate enterprise storage to manage and you can scale everything up, all at the same time.
At first I had some real questions on the support that they can give world wide as they are such a young start-up. but:
- The hardware, enterprise x86, is something you know very well.
- The base OS is VMware that you have in your datacenter, and Nutanix doesn’t modify it or write any crazy drivers for it.
- You can monitor, maintain, and mange the cluster from VMware vCenter, like U used to.
- Hardware parts replacement is pretty straightforward and includes call-home with global logistics support.
What would be the risk?