How does Hyperconverged Fit in the Enterprise? — findings from Evaluator Group Research

By , Thursday, November 30th 2017

Categories: Analyst Blogs

Tags: blog, Eric Slack, HCI, hyperconverged, infrastructure, interview, survey, thinking outside of checkbox,

Thinking_Outside_Checkbox_Banner

Hyperconverged Infrastructures (HCIs) have enjoyed significant success over the past several years, based partly on their ability to reduce the effort and shorten the timeframe required to deploy new infrastructures. HCI’s early success came largely in small and mid-sized companies, but as this technology has matured, adoption is being seen in more enterprise-level companies and in more critical use cases, to a certain extent.

Evaluator Group conducted a research study “HCI in the Enterprise” to determine how HCIs fit in enterprise environments by examining where companies are deploying or planning to deploy these products, which applications they are supporting and what benefits they expect to derive from that deployment. The study also explored the expectations of IT personnel involved in the use or evaluation of these products, including where an HCI may not be an appropriate solution. This blog discusses some the findings from this study, which included an email survey and follow up interviews with IT professionals from enterprise-level companies (over 1000 employees).

Seven out of eight enterprises we contacted are using or evaluating HCIs, or plan to in the near future. These companies told us their top IT priorities were to improve efficiency, reduce cost and increase “agility” – the ability to respond quickly to changes in infrastructure driven by their dynamic environments. Most are looking at this technology to help simplify their IT infrastructures as a way to achieve these objectives.

Simplification was a recurring theme, with infrastructure consolidation being the most common use case among the companies surveyed. These organizations are looking to HCIs as a way to reduce the number of systems in their environments and standardize on a single platform for multiple applications. This is the same approach that smaller and mid-market companies have taken for the past several years, but with a major difference.

Where smaller companies have adopted HCIs as a solution for most of their applications, enterprise IT is taking a more measured approach. They’re using HCI, but not for their most mission critical applications. The rationale seems to be that there are plenty of benefits to be had with this technology in tier 2 and tier 3 applications. When we asked why HCIs were not chosen, the reason most often given was maturity of the technology, with some concern over the fact that HCIs vendors are often start-up companies as well.

In the study we asked enterprise IT which products they were either using, evaluating or planned to evaluate. Out of the 14 products lists, the solution most often cited was VMware vSAN, with Cisco HyperFlex and Nutanix Enterprise Cloud Platform coming in second and third, respectively.

Another area we asked about was decision factors; what characteristics were the most important when comparing one HCI solution with another. Hyperconverged infrastructure products are generally “feature rich”. Most HCI vendors have also assembled a variety of models to choose from. But features and models weren’t even among the top criteria enterprise IT folks used in an evaluation. The number one characteristic was actually performance and number two was economics. What’s interesting about this is that these two characteristics are closely related.

Cost as a comparative factor is best defined in terms of ability to do useful work. In a hyperconverged cluster that work is measured by the number of virtual machines it can support. Each VM consumes storage and compute resources, more specifically CPU cycles and storage IOPS. This means HCIs with better storage performance and more CPU cores can usually handle more VMs, on a per-node basis. So HCIs with better performance often end up costing less as well. In these situations, it’s imperative that a testing suite designed for hyperconverged infrastructures, such as IOmark, be used in order to capture accurate performance data.

This study also contains detailed information on the findings mentioned above, including where and why HCIs are not appropriate, as well as input on networking, hardware platforms, the infrastructure being replaced, etc. More information and report details are available on Evaluator Group website, or contact us.


The amount and diversity of technology available in infrastructure products can be overwhelming for those trying to evaluate appropriate solutions. In this blog we discuss pertinent topics to help IT professionals think outside the checkbox of features and functionality.

Forgot your password? Reset it here.