It’s been an exciting past couple of weeks. First VMware and EMC announced their joint VxRail appliance which is, amazingly, already “industry leading.” And then yesterday Cisco, the last server vendor to fully embrace hyperconverged infrastructure (HCI), announced “HyperFlex: The next generation in complete hyperconvergence.”
But CIOs don’t care about infrastructure – hyperconverged or not. They’re looking for greater ease-of-use and reliability. They want to reduce the operational burden of the traditional datacenter so that their staff can devote their time to advancing IT support of the business rather than fighting fires.
The Broken Datacenter
The typical 3-tier (centralized storage + storage network + compute) datacenter is a mishmash of servers, SANs, NAS devices, switch fabrics and management consoles – all administered by silos of expensive IT specialists.
CIOs are no longer willing to tolerate this type of inefficiency. They are demanding the agility, responsiveness and ease of use of cloud – and if they can’t get it internally, they will absorb the higher prices and move their infrastructures to public providers.
Nutanix realized years ago that while its software-defined, share-nothing HCI architecture was a great start, the real future was in providing an enterprise cloud platform that optimizes the benefits of Cloud whether public or on-premises. Nutanix focuses on making infrastructure invisible: enabling seamless workload operation on whatever hypervisor and in whatever location organizations choose.
Eliminating complexity is the most desired attribute of what Cisco would call a “next generation” HCI platform, and yet is a difficult concept to provide. It is hardly simple, for example, when the server firmware upgrade manual is 15 pages or when a half a dozen management tools are required for an enterprise solution.
Operational simplicity including extensive automation is at the foundation of Nutanix technology. A couple of mouse clicks, for example, upgrades all of the nodes including not just the OS, but also the underlying firmware and even the hypervisor – whether vSphere, Hyper-V or Nutanix’s Acropolis Hypervisor. And these upgrades occur non-disruptively in the background on a rolling basis.
All Nutanix nodes, regardless of age or location, are all managed (and upgraded) from the single Prism Management pane-of-glass. Built-in analytics provide visibility across the entire stack.
Security is one of the most important criteria to consider when deploying datacenter infrastructure. Is security baked in or bolted on? Does the manufacturer supply STIGs (Security Technical Implantation Guides) and, if so, are they upgraded regularly?
Nutanix makes security core to everything we do. Nutanix’s full stack analytics capabilities keep the entire hyperconverged infrastructure platform baseline compliant. Pre-STIGs and automated self-healing take security to a whole new level.
Eric Hammersley, Security Architect at Nutanix recently wrote:
You should demand more from your providers. Processes that are supportable, quality assured, and not massive time sinks detracting from the actual business of the day. You pay good money for your products and shouldn’t have to spend a month tweaking and modifying configurations to get it on the network. However, that outcry from the industry is still little more than a whimper, and ownership still falls to the customer. We at Nutanix do not stand for that approach, and neither should you.
At Nutanix our Security Development Lifecycle (SecDL) ensures the product you purchase is intrinsically hardened, derived from a set of Security controls that spans as many processes and certifications as we can find, the NIST 800-53. That’s right, out of the box, already done.
Fully documented and completely transparent in its implementation, so you can verify the derived controls for yourself. That’s the responsible vendor approach, the one you deserve. Manual hardening, gone. Cyclic and monotonous application of security controls, no more. You can have your day back to do other things.
3. Virtualization Choice
Virtualization is the primary foundation of modern datacenters. But both IDC and Gartner say that over half of organizations are now running two or more hypervisors. HCI should not lock customers into a single virtualization choice.
Just as Nutanix made storage invisible, it is also making virtualization invisible. Customers can choose vSphere, Hyper-V or Acropolis Hypervisor to optimize the performance and cost of individual workloads. Nutanix even allows automated migration of hypervisors back and forth between ESX and Acropolis Hypervisor, and Hyper-V is on the roadmap.
Compression, deduplication and WAN acceleration have evolved from products to features. The hypervisor is already a non-factor in the public cloud, and Acropolis Hypervisor, which is included with every Nutanix node, is accelerating the end of the standalone hypervisor in the datacenter. Superior economics, simplicity, scalability, security, resiliency and analytics are resulting in tremendous momentum for this next-generation hypervisor.
4. Cloud Choice
Nutanix includes the Acropolis App Mobility Fabric which enables customers to back up seamlessly to either AWS or to Azure. Spinning up the VMs for disaster recovery purposes is next. Eventual full convergence with private and public clouds will minimize customer costs by seamlessly running elastic workloads in public clouds and predictable workloads on-premises.
Nutanix wields support as a competitive advantage. We boast an industry-leading 92 Net Promoter Score. Nutanix support covers the entire infrastructure stack: Compute, Storage and Virtualization. Nutanix even provides support for business critical applications such as SQL & Exchange.
Nutanix owns the problem regardless of whether the issue is believed to be a 3rd party responsibility or not. Nutanix SREs will arrange and actively participate in multi-way calls between vendors until the customer’s problem is resolved. While Nutanix SREs are highly skilled and experienced, they also can escalate to multiple CCIE, VCDX, NPX and CTPs.
A Strategic Approach to Purchasing IT Infrastructure
Due to the scalability and other limitations of 3-tier infrastructure, IT organizations have been used to thinking tactically about purchase decisions – buying compute and storage components separately as required. We see this frequently during our TCO/ROI analyses when customers ask us questions such as, “What is your cost per GB?”, or “What does it cost to buy a Nutanix pod – because that’s how we buy infrastructure?” While perfectly legitimate questions in a 3-tier environment, they make no sense in a web-scale world.
The whole Converged Infrastructure genre, despite lacking any real innovation or, for that matter, any actual infrastructure that is converged, has done well because it helps mitigate the challenges of standing up and supporting/troubleshooting 3-tier infrastructure. But converged infrastructure solutions still require separate tiers of servers and proprietary (hardware-defined) storage, with each tier managed separately.
HCI is barreling toward the chasm crossing. In addition to the new products by Cisco, EMC and VMware, HPE announced an integrated solution with Azure a few months ago. Dell and Lenovo both sell Nutanix-powered HCI appliances.
It behooves all organizations considering their IT futures to now take an honest look at replacing their 3-tier environments with a combination of hyperconverged and public cloud.
Thanks to @joshodgers, @cakeis_not_alie, @sudheenair and @JonKohler for edits and suggestions.