Last week, HP Enterprise announced its agreement to acquire Hyperconverged Infrastructure (HCI) vendor Simplivity for $650M. You can read the story here.
Big news? I guess it depends on how you look at it. The HPE acquisition had been rumored for some time. So, it wasn’t a big surprise, but it is still big news. It’s yet another validation of the importance of the growing hyperconverged infrastructure market, and more importantly, it adds another prominent name to the growing list of mainstream IT system players, who are lining up to do battle in the HCI arena.
Now the question is how does Dell/EMC, IBM, Lenovo, Nutanix and Cisco respond, and what is the broader HCI strategy inside of HPE? We all know the low and mid-range storage business is shifting to HCI away from Fibre-channel. Does HCI stay in the lower end and single workload (VDI) environments or move more mainstream mixed workloads and up market?
So, what’s all the fuss about HCI? Why is HCI the next great thing? Well, it depends on how you look at it. From a market perspective, HCI continues to grow at the expense of the traditional server and storage market share due to its modular, scale-out architecture. (This large and growing opportunity is why the market is becoming crowded.) From a customer perspective, HCI offers a very practical, cost-effective approach to deploying IT infrastructure. It enables customers to grow their infrastructure incrementally over time without having to rip and replace their existing datacenter assets. Ultimately, HCI gives customers much better agility to quickly react to change, and HCI can provide the infrastructure flexibility needed to support new consumption models for IT consumers, like HCI-as-a-service.
According to Gartner, “Ultimately, the underlying infrastructure will disappear to become a malleable utility under the control of software intelligence and automated to enable IT as a service (ITaaS) to business, consumer, developer, and enterprise operations.”
As I highlighted in my last blog post, delivering anything “as-a-service” requires the control and agility achieved through software. In other words, the modern data center infrastructure needs to be “software defined.” Innovative software, running on industry-standard hardware, represents the industry’s direction for the Software Defined Data Center (SDDC). We see this from the web-scale guys, like Google and Amazon, who run their datacenter infrastructure software on standardized, commodity hardware to significantly reduce complexity and costs, and improve scale and we see the HCI vendors deploying their software stacks on the same.
So far, HCI vendors have defined converged software stacks for software-defined compute and storage, but there is still a critical missing piece — the network. In lieu of the missing network component of HCI, vendors have offered best-practice guidelines for retrofitting existing, traditional network infrastructure to support HCI. These guidelines are based partly on the deficiencies of traditional (20-year old) “hard-wired” networks, and therefore recommend the deployment of several, separate, physical networks to ensure proper workload path isolation and achieve predictable performance and security. This vendor-recommended networking approach is complex, costly, and inefficient; it simply doesn’t match up with the agility, ease-of-use, and cost benefit profile of HCI.
And we’re not the only ones that think so. In a recent newsletter, entitled, “Gartner Newsletter: Optimizing Hyperconverged Infrastructure with Plexxi,” Gartner talks about the criticality of the network in the HCI equation and points to software-defined networking (SDN) as critical to the HCI vision. As Gartner points out in their report, “Most HCIS deployments today have largely ignored the network. Mixing user access, internode application traffic, internode storage traffic, VM mobility, and storage access traffic on a single network switch without bandwidth control can lead to unpredictable performance and system stability, and compromised data integrity.”
Basically, what Gartner is saying is that HCI introduces new network workloads that are placing new requirements on the network, requirements that traditional networks were not designed to address. Most early HCIS deployments consist of a handful of nodes “often hosting less-demanding workloads such as virtual desktop infrastructure (VDI), application development, and remote office/branch office infrastructure.” It’s when these HCIS clusters grow and begin to support applications with more demanding I/O workloads that network traffic challenges become real problems, unless the network component is addressed. In fact, according to Gartner, the I/O loads alone “will increase by 10x” with HCIS clusters.
Plexxi is unique in the industry and delivering today, the Hyperconverged Network for HCI. Plexxi’s HCN offers the level of control and workload isolation crucial to HCI. With Plexxi, organizations can deploy a single IP network on industry-standard hardware, that supports all HCI and application workloads. Network administrators can define workload policies, allocate path bandwidth, and isolate workload traffic across the Plexxi fabric, eliminating the need to deploy separate IP networks to meet performance, availability, and security SLAs. As a software-defined network architecture, Plexxi provides the management interfaces and application APIs needed to build an agile datacenter network.
The old battlefront between stove piped technology vendors offering discrete compute, storage, and networking solutions based on purpose-built hardware is giving way to a new battlefront, the Software Defined Data Center, in which major innovations are in software running on industry-standard, commodity hardware. HCI solutions, based on software-defined compute, storage, and networking, are poised to change the datacenter landscape as we know it. Plexxi’s hyperconverged network is software-defined, uses industry-standard hardware, and is available today. We’re Simply a Better Network for HCI.
 Gartner, Press Release – 5 May 2016