Author: Client Executive – Jamie Halliday
Why Hyperconvergence is Revolutionary
With the recent emergence of hyperconverged infrastructure, a cloud enterprise solution, age old IT thinking is being challenged in almost every aspect. It’s now possible to have both the security of on premise infrastructure, as well as the flexibility and disaster recovery capabilities of cloud solutions. This article will seek to explore the benefits of the new technology, and how it’s slowly but surely redefining the infrastructure landscape.
- Legacy three-tier infrastructure
- Over-provisioning vs under-provisioning
- Infrastructure Lifetime
- Changing the infrastructure game
- What does this mean for our infrastructure team?
- Why simple is better
- Is it time to review your infrastructure stack?
- Take the Next Step
Legacy three-tier Infrastructure
Three-tier infrastructure has been the cornerstone of operations in IT circles for the last several decades, with many backing the necessity of standalone hardware. The majority of medium to large businesses today have some form of infrastructure team in-house to monitor and maintain the internal tech of the office, enabling day-to-day activities to take place.
Such is the passion for legacy infrastructure that conversations on expenditure are disregarded immediately, and the examination of alternatives is tantamount to treason. Well, perhaps not treason, but the opinion remains that while three-tier may be costly, it is a necessary investment for the business. However, is it really a genuine, unavoidable necessity?
I’ll save you the trouble of answering that question for yourself. No, it’s no longer necessary now that there’s hyperconverged infrastructure. Before we jump into the new technology though, let’s look at traditional, converged infrastructure.
Over-provisioning vs under-provisioning
Perhaps the largest challenge that faces IT departments running three-tier infrastructure is ensuring that the organisation is sufficiently able to handle all the business’ activities. The age-old question of “Do we have enough storage and compute not only for today, but also for 6 to 12 months down the line?” starts to surface every time a refresh is on the horizon.
To reduce the risk of being short handed on either of these requirements, IT departments may “over-provision” their infrastructure. This is simply the process of purchasing more than needed during an infrastructure refresh with the expectation of “growing” into the need. While this approach will certainly help to reduce the strain on infrastructure, it creates the problem of idleness and asset underutilization. Scenarios unfold in which infrastructure is needed at a high-level for one-off or rare instances, but otherwise is left to metaphorically “twiddle its thumbs”, for large periods.
Think of a married couple looking to buy a new home to begin a family. It’s unlikely that they would look to buy a 6-bedroom house when they currently don’t have any children. Yes, they may grow into the need eventually, but there are several rooms that are not likely to be used for some time.
The reverse is also a problem. When IT departments under-provision infrastructure, it opens up the possibility of coming up short when it comes to crunch time, leaving the organization scrambling to fix the situation. While it is true that your infrastructure may be running closer to true efficiently most of the time, under-provisioning can lead to stoppages, which can be seriously problematic if a client project is dependent on performance and a tight schedule.
We can go back to the family buying a house analogy. You wouldn’t buy a one bedroom house if you’re looking to start a family soon, it just doesn’t make sense.
Both over-provisioning and under-provisioning are situations that are costly to maintain, particularly given the limited lifespan of infrastructure.
Naturally, nothing lasts forever. Regardless of whatever technology is in play, it will eventually break down and fail.
The additional challenge to three-tier however, is the way in which it is constructed. It’s made up of many different interchangeable parts. Different bits have different lifetimes, which can lead to a constant need to review and replace. It’s rarely a hassle free process for IT departments.
Changing the infrastructure game
Can anything be done about the challenges around infrastructure provisioning and the limitations of hardware lifespan? The answer is yes, it comes in the form of hyperconverged infrastructure.
Hperconverged infrastructure (HCI) is a new development in the market that takes server, network and storage, packaging it all nicely into one box. No longer is it necessary to manage the multiple layers and complexities of three-tier infrastructure.
This sandwiching of all these functions not only reduces the technical complexity, but also increases the accountability. The three-tier approach would have you jumping through a number of hoops to speak to all of your different infrastructure vendors when something goes wrong. Even then, pinpointing the problem could become more difficult as a result of “finger-pointing”. With HCI, you take multiple vendors out of the equation, leaving a single vendor to talk to if issues do arise.
From a provisioning point of view, HCI is made up of nodes, with each one containing storage, server and network functions. These nodes can be thought of as Lego blocks, allowing you to build and scale quickly and easily. This added flexibility allows for rapid deployments in a matter of hours, rather than days or weeks, as well as the ability to linearly scale up or down, depending on your requirements. Because of this linear ability to scale, there is no true theoretical limit to HCI provisioning.
A regular challenge to IT departments is the predictability of workloads. What do you do if workloads are erratic and highly variable? The answer is public cloud.
IaaS (infrastructure-as-a-service), PaaS (platform-as-a-service) or Saas (software-as-a-service) solutions can provide much needed flexibility to handle unpredictable workloads. For example, an organisation may build an application on the public cloud, and then when the workload becomes stable or predictable, they can bring it back in-house.
If you are familiar with public cloud, you will most probably know that renting public cloud is nowhere near as economical as owning your infrastructure when you know it is going to stay in the same place for some time.
HCI gives you the ability to own your infrastructure on premises, with the flexibility to “slingshot” workloads in and out of the public cloud easily when required. Think of HCI as a powerful tool to help utilise public cloud in a way that makes the most sense.
What does this mean for our infrastructure team?
With the reduction in complexity afforded by HCI, less time and energy is required to keep your day-to-day infrastructure functioning. While some may see this is a potential headcount reduction strategy, consider the investment in HCI as more of a headcount re-purposing opportunity. Are there projects that are on hold due to an overloaded infrastructure team? Are there development opportunities that have been left on the back burner? At the end of the day, it will benefit the IT staff as well, as I’m sure we can agree that nobody really wants to be up till 2am working to fix something that shouldn’t have gone wrong in the first place.
Ask yourself this question, “Is my IT department generating new value and improving office performance in a consistent way, or are they simply fighting the fires as they appear?”
Why simple is better
The direction of technology has always focused on improvement and automation. It is all about making life easier and simpler, not harder and more complex.
Why can’t IT infrastructure be the same? Perhaps it’s the false perception that simplicity implies inferiority. This could not be further from the truth. While hyperconverged infrastructure may be easier to understand and use than three-tier, it would be wrong to assume that it lacks its functionality and power.
In fact, HCI actually allows for better all-round performance because of its use of nodes. The VMs (virtual machines) on every node are serviced by what is known as a “controller virtual machine”. This allows for linear scaling of not only capacity, but performance and resiliency as well.
Simple management also means simple upgrades. Imagine being able to perform a disruption free upgrade at the single click of a button. All this is possible with HCI.
Is it time to review your infrastructure stack?
Ultimately, HCI is built for the future. It has emerged from the growing need for flexibility, agility and scalability in businesses today.
Do any of the issues above resonate with your IT infrastructure environment? Perhaps it’s time to review the areas that challenge your business, and assess how HCI might be able to help.
Take the Next Step
Nutanix enterprise cloud is a HCI solution that natively converges storage, compute and network into a single unit. Speak to us today about how Nutanix can consolidate your IT infrastructure, reduce your TCO and improve your performance.
Want us to cover a particular topic? Leave a comment or email us at firstname.lastname@example.org.