Considering Hyper-converged Infrastructure? Ask Yourself These Three Questions
As hyper-converged IT infrastructures become increasingly mainstream, enterprises must brush up on the approach lest they fall behind industry standards.
As recently as a decade ago, enterprises relied almost exclusively on strictly siloed IT infrastructures — not by choice, but by necessity. IT administrators would create one team to manage server-hosted compute functions, another to manage storage operations, and yet another to manage networking. Thanks to the rise of the cloud and, more recently, converged and hyper-converged infrastructures (HCI), this need no longer be the case.
While the hype surrounding HCI has blurred its definitional boundaries — especially along its border with converged infrastructure — it hasn’t diminished the immense value HCI has to offer to savvy enterprise IT teams. Defined by TechTarget as “a software-centric architecture that tightly integrates compute, storage, and virtualization resources in a single system that usually consists of x86 hardware,” HCI provides enterprise IT teams with easier scaling, streamlined management, and a lower total cost of ownership than traditional IT infrastructures.
However, despite their reputation as “mini data centers in a box,” realizing the full potential of HCI nodes takes more than just “plugging and playing,” especially over the long term. As such, enterprises that are considering HCI should ask themselves the following three questions before making the switch.
1. How flexible is the hyper-convergence software I’m considering?
Hyper-convergence software can be purchased in a bundle with commodity hardware (like x86) — that is, as a mini data center in a box — or as a software-only layer that can be installed on top of an enterprise’s existing hardware.
While resource consolidation is the primary value-add of HCI, enterprises should steer clear of hyper-convergence software that is tied too closely to specific underlying hardware. Most enterprise infrastructures feature components from a variety of manufacturers (and a variety of “generations”), and an enterprise’s hyper-convergence software must be flexible enough to function across all of these technical variations.
This is not to say that hyper-convergence software should never be purchased in a bundle, only that enterprise stakeholders should pay close attention to interoperability when selecting their software. If, for instance, the vendor from which you’ve purchased the majority of your existing IT components offers a mature hyper-convergence solution, opting for this solution is a surefire way to ease interoperability concerns. (Though, as always, beware vendor lock-in.)
2. Am I taking a long-term view of my enterprise’s needs?
The speed of technological development makes genuine future-proofing all but impossible in the IT space, but enterprises must nevertheless do their utmost to ensure that the products they purchase align as closely as possible with their long-term needs. Many vendors offer guarantees that they will support their HCI offerings for the duration of their viable life, giving enterprises the peace of mind that comes from knowing that they will continue to have access to incremental server, storage, and networking capacity for the foreseeable future.
These guarantees notwithstanding, an enterprise should never calibrate the purchase of any tech — let alone HCI components — based on its current (or even short-term) needs. As a rule of thumb, an enterprise should forecast its workload requirements and application data growth two to three years into the future, and look for an HCI system that is capable of handling these demands right out of the box.
3. Can my IT staff handle the transition to — and subsequent management of — a hyper-converged infrastructure?
As previously mentioned, one of the clear selling points of HCI is that it can streamline an enterprise IT team’s management tasks. That said, the extent to which this actually occurs will depend upon how closely hyper-convergence software is tied to its underlying hardware. IT generalists can easily manage tightly bundled “black box” HCI nodes, eliminating the need for an enterprise to keep highly trained specialists on staff. Conversely, managing hyper-convergence software that is decoupled from specific IT components requires just as much — if not more — expertise than managing a traditional IT infrastructure.
In short, enterprises that need to be able to exercise close control over the configuration of their IT resources and the granular provisioning of their various software systems shouldn’t rely on pre-packaged HCI nodes. As a result, they need a large, experienced IT staff to ensure that their real-time resource orchestration goes off without a hitch.
For many enterprises, building such a team can be prohibitively expensive. That’s where a managed IT services provider like Turn-key Technologies (TTI) comes into play. At TTI, we’ve been helping enterprises optimize their IT infrastructures for nearly three decades, and our technicians are capable of getting the most out of traditional, converged, and hyper-converged infrastructures alike.
Despite its relative youth, HCI has already become an important part of a cutting-edge enterprise IT operation, and a strong managed services partnership is often the best way for an enterprise to catch up to the industry vanguard.
Want to learn more about hyper-converged infrastructure? Check out our latest digital story!
By Craig Badrick