Nutanix Cloud Infrastructure (NCI) Reviews

Originally posted at http://vcdx133.com

In my opinion, Hyper-Converged Infrastructure (HCI) is the future of Private and Hybrid Cloud infrastructure. If you are designing a greenfield data center and HCI is not on your list for serious consideration, then it should be.

With that being said, every technology has its Pros and Cons; nobody rides for free. The advantages of HCI are many and outweigh the disadvantages, here is what you need to watch out for:

• People – If your server virtualization, network and storage teams operate in silos and are constantly at war with each other, then HCI is probably not for you at this time. Successful HCI projects are built upon a very close collaboration between these teams. In fact, it makes more sense to merge these three teams into one “Enterprise Infrastructure” team. It is also very important to cross-skill these team members and let them evolve into “Enterprise Architects”, “Enterprise Administrators” and “Enterprise Operators”. However, make sure you keep your Backup/Recovery/Archive responsibilities separate (see RBAC point below).
• Data Center Facilities – A data center full of legacy, 3-tier infrastructure is not the same as one packed with HCI. The resource density ratio is around 4-8 to 1 depending upon your current legacy infrastructure. You need to design for 25+kW racks with a matching cooling system. If you use a traditional, legacy data center (designed for 5-8kW per rack), then you will have problems down the road (hot spots and power exhaustion).
• Switch Fabric – By moving to HCI, you need a scalable LAN fabric that provides non-blocking throughput for East-West traffic. Legacy network switch design (Core – Distribution – Aggregation – Access layers) is not going to cut it for large scale HCI, which is optimised for North-South traffic. You may get away with it initially, however you will need plans to migrate to a non-blocking leaf and spine switched LAN. HCI has made Fiber Channel infrastructure obsolete, but the same principles that drove SAN design now apply to your LAN with the move to IP storage.
• Controller VM – The storage processor of legacy storage arrays has now become a virtual appliance running on the host itself. Make sure your administration/operations staff, Standard Operating Procedures and monitoring systems understand the importance and give it the respect it deserves. The current version of NOS with ESXi still allows vSphere administrators to modify the CVM (Nutanix Acropolis does not allow this for CVMs with Nutanix KVM). For example, an untrained vSphere Administrator powers off all Nutanix Controller VMs and reduces the RAM from 24GB to 8GB to provide additional resources for adding new VMs across the entire cluster.
• Role Based Access Control – When I consider failure scenarios for Business Continuity and Disaster Recovery, my nightmare risk is not a natural disaster, but the disgruntled rogue administrator, who has all of the keys to the kingdom, taking out every system. With HCI and the “Enterprise Administrator”, this risk is compounded. So it is very important to separate the administration/operations responsibilities for operational data and backup/recovery/archive. This way if either one is wiped out across all data centers, you still have the other to recover from. Apply this concept to physical data center security as well.
• Data Locality and the Working Set – “Data Locality” is the amount of local storage resources (capacity and performance) presented via the Controller VM to the Hypervisor for serving your virtual workloads. The “Working Set” is the active footprint (capacity and performance) of those virtual workloads. As an organization (architects, administrators and operators), you need to make sure that the “Working Set” of your virtual machines have the optimum fit with respect to the “Data Locality” of each node in your HCI solution. Nutanix XCP has many different models, you need to make sure you select the correct fit for your needs.
• Processes and Procedures – In my opinion, Hyper-Converged Infrastructure (HCI) is the future of Private and Hybrid Cloud infrastructure. If you are designing a greenfield data center and HCI is not on your list for serious consideration, then it should be. With that being said, every technology has its Pros and Cons; nobody rides for free. The advantages of HCI are many and outweigh the disadvantages, here is what you need to watch out for:
• People – If your server virtualization, network and storage teams operate in silos and are constantly at war with each other, then HCI is probably not for you at this time. Successful HCI projects are built upon a very close collaboration between these teams. In fact, it makes more sense to merge these three teams into one “Enterprise Infrastructure” team. It is also very important to cross-skill these team members and let them evolve into “Enterprise Architects”, “Enterprise Administrators” and “Enterprise Operators”. However, make sure you keep your Backup/Recovery/Archive responsibilities separate (see RBAC point below).
• Data Center Facilities – A data center full of legacy, 3-tier infrastructure is not the same as one packed with HCI. The resource density ratio is around 4-8 to 1 depending upon your current legacy infrastructure. You need to design for 25+kW racks with a matching cooling system. If you use a traditional, legacy data center (designed for 5-8kW per rack), then you will have problems down the road (hot spots and power exhaustion).
• Switch Fabric – By moving to HCI, you need a scalable LAN fabric that provides non-blocking throughput for East-West traffic. Legacy network switch design (Core – Distribution – Aggregation – Access layers) is not going to cut it for large scale HCI, which is optimized for North-South traffic. You may get away with it initially, however you will need plans to migrate to a non-blocking leaf and spine switched LAN. HCI has made Fiber Channel infrastructure obsolete, but the same principles that drove SAN design now apply to your LAN with the move to IP storage.
• Controller VM – The storage processor of legacy storage arrays has now become a virtual appliance running on the host itself. Make sure your administration/operations staff, Standard Operating Procedures and monitoring systems understand the importance and give it the respect it deserves. The current version of NOS with ESXi still allows vSphere administrators to modify the CVM (Nutanix Acropolis does not allow this for CVMs with Nutanix KVM). For example, an untrained vSphere Administrator powers off all Nutanix Controller VMs and reduces the RAM from 24GB to 8GB to provide additional resources for adding new VMs across the entire cluster.
• Role Based Access Control – When I consider failure scenarios for Business Continuity and Disaster Recovery, my nightmare risk is not a natural disaster, but the disgruntled rogue administrator, who has all of the keys to the kingdom, taking out every system. With HCI and the “Enterprise Administrator”, this risk is compounded. So it is very important to separate the administration/operations responsibilities for operational data and backup/recovery/archive. This way if either one is wiped out across all data centers, you still have the other to recover from. Apply this concept to physical data center security as well.
• Data Locality and the Working Set – “Data Locality” is the amount of local storage resources (capacity and performance) presented via the Controller VM to the Hypervisor for serving your virtual workloads. The “Working Set” is the active footprint (capacity and performance) of those virtual workloads. As an organization (architects, administrators and operators), you need to make sure that the “Working Set” of your virtual machines have the optimum fit with respect to the “Data Locality” of each node in your HCI solution. Nutanix XCP has many different models, you need to make sure you select the correct fit for your needs.
• Processes and Procedures – Moving from legacy, 3-tier infrastructure to HCI is a big change, so do not underestimate or ignore the imperative to update all of your processes and procedures. HCI will simplify and improve your infrastructure, consequently simplifying your operational procedures, but you will need to change how you do things with respect to people, process and technology.

What is our primary use case?

How has it helped my organization?

What needs improvement?

For how long have I used the solution?

What do I think about the stability of the solution?

What do I think about the scalability of the solution?

How are customer service and support?

How was the initial setup?

What's my experience with pricing, setup cost, and licensing?

What other advice do I have?

Which deployment model are you using for this solution?

What is our primary use case?

What is most valuable?

What needs improvement?

For how long have I used the solution?

What do I think about the stability of the solution?

What do I think about the scalability of the solution?

How are customer service and support?

Which solution did I use previously and why did I switch?

How was the initial setup?

What's my experience with pricing, setup cost, and licensing?

What other advice do I have?

Which deployment model are you using for this solution?

What is most valuable?

How has it helped my organization?

What needs improvement?

For how long have I used the solution?

What do I think about the stability of the solution?

How is customer service and technical support?

How was the initial setup?

What's my experience with pricing, setup cost, and licensing?

Which other solutions did I evaluate?

What other advice do I have?

What is most valuable?

How has it helped my organization?

What needs improvement?

For how long have I used the solution?

What do I think about the stability of the solution?

What do I think about the scalability of the solution?

How are customer service and technical support?

Which solution did I use previously and why did I switch?

How was the initial setup?

What's my experience with pricing, setup cost, and licensing?

Which other solutions did I evaluate?

What other advice do I have?

Originally posted at www.storagegaga.com/dont-get-too-drunk-on-hyper-converged/

I hate the fact that I am bursting the big bubble brewing about Hyper Convergence (HC). I urge all to look past the hot air and hype frenzy that are going on, because in the end, the HC platforms have to be aligned and congruent to the organization’s data architecture and business plans.

The announcement of Gartner’s latest Magic Quadrant on Integrated Systems (read hyper convergence) has put Nutanix as the leader of the pack as of August 2015. Clearly, many of us get caught up because it is the “greatest feeling in the world”. However, this faux feeling is not reality because there are many factors that made the pack leaders in the Magic Quadrant (MQ).

First of all, the MQ is about market perception. There is no doubt that the pack leaders in the Leaders Quadrant have earned their right to be there. Each company’s revenue, market share, gross margin, company’s profitability have helped put each as leaders in the pack. However, it is also measured by branding, marketing, market perception and acceptance and other intangible factors.

Secondly, VMware EVO: Rail has split the market when EMC has 3 HC solutions in VCE, ScaleIO and EVO: Rail. Cisco wanted to do their own HC piece in Whiptail (between the 2014 MQ and 2015 MQ reports), and closed down Whiptail when their new CEO came on board. NetApp chose EVO: Rail and also has the ever popular FlexPod. That is why you see that in this latest MQ report, NetApp and Cisco are interpreted independently whereas in last year’s report, it was Cisco/NetApp. Market forces changed, and perception changed.

The most glaring gap of the Gartner MQ is, it does not measure how good each technology is against the competitors, and therefore, boxes like Nutanix and Simplivity, and also the motley crew of so-called EVO:Rail partners, are simply betting on how good they pitch to get into the MQ Leaders Quadrant as part of their agenda.

It is not common that a vendor would do a bake-off against a competitor because one might fall flat on his/her face if they lose the bake-off. Recently we witnessed the big shouting match between Nutanix and VMware, each trying to out-boast the other with claims and counter claims of how good their performance was. Here is VMware’s Chuck Hollis’ piece and here is Nutanix’s piece (part 4 anyway). It was hounding the blogosphere with so much *bleep* that it felt like a game of Kabaddi. (Look up the game of Kabaddi).

The dare, double-dare and triple-dare went on for a few weeks between VMware and Nutanix until Storage Review, an online “independent” storage articles and news aggregator, probably got sick of this undignified spat. Storage Review ran the Nutanix’s tests (and blogged about it) and came up with some unexpected results. Nutanix got disillusioned and started to dictate what should be in the performance tests and went through some “innocent” and “hilarious” acts to justify the results. You can read about them here. Page 2 of that article started the “We ask you not to …” comments to Storage Review, which I thought was really funny.

That’s just it. The Hyper Converged vendors spend a lot of the resource and marketing on hypping up performance, and little of everything else. It is true that storage performance is needed for 25-30% of the active data. However, in any organization, we know very well that 70-75% of the data in the entire data lifecycle is non-active, Tier-3 or archived data. Hyper Converged systems and platforms ignore this passive data space. They do not play with inactive data landscape because the $/GB or $/TB is too expensive.

Furthermore, if we look at the entire HC thingy more close, it is just Software-Defined because it is just a wonderful piece of proprietary software running on an ODM (original design manufacturer) x86 platform. EMC VSPEX is rumoured to be running on the Quanta systems and Nutanix on Supermicro. Why do we need to pay a premium price for a piece of whitebox hardware?

We must overcome the hype and read beyond the messages that vendors, partners who tend to oversell their technology. We must look at what is required in our business, in our operations and look at our data requirements. I constantly share that we must look at our data landscape and use the 7 points that I always use to consider a technology – Availability, Performance, Protection, Accessibility, Management, Security and Compliance. Is the HC technology and the solution relevant to your organization’s business?

So, if you want to Hyper Converged, please Hyper Converge responsibly. Your data landscape architecture will thank you for it.

Originally posted at vcdx133.com.

This post provides a Tech101 breakdown of the Nutanix Virtual Computing Platform. If you have heard the buzz-words “Nutanix”, “Web-Scale” and “Hyper-Convergence” and want to learn more about it, this post is for you.

The Nutanix Virtual Computing Platform (VCP) is a Hyper-Converged Infrastructure solution. Termed the “iPhone” of infrastructure, what is the big deal?

Short answer: Ease of use, fast deployment, linear scaling of performance with capacity and richness of product features.

Why should you care? If you live in the world of monolithic and vertically scaled storage, you may have experienced the initial, massive cost of purchasing your storage solution with a 5 year ROI. Then comes the honey-moon period where you experience amazing performance as workloads are moved to the new array. Well before the 5 year ROI mark, you will probably experience performance degradation as the capacity of the array is consumed. Which then initiates discussions with your storage vendor about how you need to spend another boatload of money adding additional controllers, SSDs and disks to the array.

Furthermore, think about how long it takes for the storage team, the server team and the network team to agree on a particular design for a project and the time it takes to order, deliver, install, configure and test that solution. Have you ever asked yourself, “There must be an easier way to do this?”

Enter Hyper-Convergence, where you can now buy individual blocks of infrastructure at an entry price and quickly scale-out your solution (performance and capacity) one unit at a time as you grow.

What is it? Nutanix has taken the “Web-Scale” infrastructure model of Google and Facebook and made it available to the Enterprise. You buy a Nutanix Block (rack mounted chassis) that contains Nutanix Nodes (hosts), storage, power and cooling. This creates a Nutanix Cluster (Software-Defined Storage), which are comprised of Controller Virtual Machines (CVM).

How does it work? The Controller Virtual Machines connect directly to the physical SCSI Controllers on each node (Pass-Through mode) and the CVMs are configured to create a “Nutanix Distributed File System (NDFS) Cluster” (storage pool) via the 10GbE network. All of the pooled storage is classified as local or remote from the point of view of each CVM and VM data is protected based upon a Replication Factor (RF) policy of two or three, which replaces the concept of RAID. RF-2 means a local and remote copy is maintained and RF-3, a local and two remote copies of said data.

Where is the magic? Since each node has its own Controller VM, SSDs and HDDs, NDFS can provide linear scaling of capacity and performance for locally stored data. When a CVM goes down, a remote CVM will takeover serving data I/O to the local hypervisor. When the local CVM comes online, it resumes service. This fundamental concept is what allows Nutanix to provide a “rolling upgrade” of the Nutanix Operating System (NOS). What is the caveat? During periods of CVM failure, local workloads that access remotely stored data will experience degraded performance until the local CVM is restored.

The major components of the Nutanix VCP solution are:

• Nutanix Node – a single host with CPU, RAM, SCSI Controller and Network that resides in a Nutanix Block
• Nutanix Block – rack mounted chassis that contains nodes, storage, power and fans
• Prism – User Interface and APIs
• Prism Central – aggregator of separate Prism Clusters to a single UI
• Cluster Virtual Machine – Software-Defined Storage that supports NFS, iSCSI or SMB3
• CVM operating system is known as “NOS” – Nutanix Operating System
• CVM file system is known as “NDFS” – Nutanix Distributed File System
• Hypervisor – VMware vSphere, Microsoft Hyper-V and KVM supported
• OEM Hardware – Supermicro and Dell with PCIe SSD and HDD
• Licencing – Starter, Pro & Ultimate

It comes in a number of OEM Hardware flavours (Dell and Supermicro) with the following uses:

• ROBO, Test & Development – NX-1000, Dell XC 720xd A5
• EUC/VDI – NX-3000, NX-7000 (GPU & PCoIP Offload), Dell XC 720xd B5/B7
• Compute Intensive – NX-3000, Dell XC 720xd B7
• Data Intensive – NX-8000, Dell XC 720xd C5/C7
• Business Critical & Mission Critical Apps – NX-9000 (All Flash)

The major features of the Nutanix Operating System (current version NOS 4.1):

• One-Click Upgrades of NOS and Hypervisor (rolling upgrade)
• Data Protection Cloud Connect (to AWS)
• Data At Rest Encryption
• Metro Availability (<5ms RTT)
• Deduplication
• Compression
• Data Tiering
• Snapshots and Clones
• Advanced Data Protection

Alternatives

• Server-Side Flash-Cache Acceleration – VMware vFRC, PernixData FVP, EMC XtremSF/XtremCache, etc.
• Converged Infrastructure Solutions – NetApp FlexPod, VCE vBlock, HP ConvergedSystem, Dell, etc.
• Other Hyper-Converged solutions – VMware EVO:Rail, Maxta, SimpliVity OmniCube, Scale Computing, Pivot3, etc.

What is our primary use case?

What is most valuable?

What needs improvement?

For how long have I used the solution?

What do I think about the stability of the solution?

What do I think about the scalability of the solution?

How are customer service and technical support?

Which solution did I use previously and why did I switch?

How was the initial setup?

What about the implementation team?

What's my experience with pricing, setup cost, and licensing?

What other advice do I have?

Which deployment model are you using for this solution?