Making Sense of Hyperconverged Infrastructure
Your Stepping Stone to the Cloud
The term hyperconverged infrastructure (HCI) was first coined in 2012. Since then, it has grown exponentially, with Gartner predicting a 48% compound annual growth rate over the next 5 years.
That’s not surprising when you consider HCI has been shown to improve operational efficiency, scalability, and data efficiency while reducing costs.
We asked Josh Moore, one of ITP’s senior solutions architects, to talk about what it is, what it does, and how businesses can determine if HCI is the right option for them.
How We Got to Hyperconverged
Back in the day, a lot of companies used physical servers, localized storage, and switches to connect everything into the network.
Then it moved to a three-tier architecture where you had the compute power plugged into a centralized storage area network that had all of these disc arrays to power your application workload. In that three-tier setup, you had multiple vendors – maybe you had HP servers, a different manufacturer for your SAN, and then the switching architecture.
Hyperconverged connects storage, processor, and memory all into a big pool that you can allocate virtual machines onto. It’s a good way to simplify that architecture but get some of the benefits of enterprise resiliency.
In any infrastructure, you have points of failure. What you want to have is redundancy in connectivity between the pieces in your network to make sure that if any one piece fails that you’re not “down.”
Unlike the three-tiered system with multiple vendors, hyperconverged mashes that all into one configuration, one setup. Why is that good? More simplification, better management dashboards, better transparency into your network, and a single throat to choke for support.
4 Ways HCI Helps Businesses
Support – Hyperconverged infrastructure means being able to work with a single vendor to troubleshoot all that compute power, all that storage, connectivity, etc.
Transparency – You experience real transparency into the performance of the architecture. Because the infrastructure is all in one place, you have better reporting and management capabilities. You can easily see if a server is reaching maximum memory, or processor capacity, or disc.
Disaster Recovery – A lot of HCI vendors have DR options where you can place another node at another site to ensure that if there’s a catastrophe (tornado, flood), that other data center is able to continue powering your applications so there’s a level of business continuity in place.
Gateway to the Cloud – Most HCI providers know that everybody wants to move to the cloud. So they offer plugins to move workloads between the HCI and a cloud service like Microsoft Azure. There are also lots of tools to help you look at the cost component of running a workload on-premise vs. in the cloud. HCI is really one step closer to the cloud. In fact, many customers call it their private cloud because it’s very similar. You get out of the hardware game, you get out of managing a lot of those individual components, and you’re just focused on “what servers do I need to power the application workload to support the business?”
How Do I Know if HCI is Right For My Business?
The entry into hyperconverged is different for every industry vertical. That said, the ideal hyperconverged customer has a workload where they need uptime and resiliency. At the same time, they want to simplify the operational management of their servers.
In many cases, what eventually drives a business to make the leap to HCI is one of two things. One, they’re at the point where their system equipment has reached the end of its life and needs to be replaced. Or two, they’ve experienced business growth to the point where their current infrastructure isn’t supporting the business anymore.
I recommend researching it for yourself and then talking to an IT provider with experience in hyperconverged.
I’m Still Hesitant to Try HCI. Am I Alone?
In my experience, I’ve seen there’s still a preference among companies to have their own hardware. A lot of companies are looking to move to the cloud, but there’s still a great number of legacy applications that dictate that you need to have that app workload on-premise.
At first, many businesses don’t understand the intricacies of a three-tiered system vs. hyperconverged. And that’s ok. But once you kind of “get under the covers” and start to talk about simplifying the infrastructure as well as leveraging hyperconverged as a gateway to the cloud, then it becomes a very compelling case for people.
One Last Thing
Hyperconverged is a great stepping stone to cloud migration for people who need that on-prem workload. Taking that hybrid approach for a lot of businesses is the right idea. Instead of just jumping into the cloud, you can put a toe in and test it out with HCI.
To learn more, visit our Hyperconverged Infrastructure page.
What Is Hyperconvergence? (Network World)