26 Jan Hyperconvergence Pros and Cons
The debate about hyperconvergence is, for all intents and purposes, over. No longer is there a question about whether hyperconvergence will take off: it has. No one wonders if it will find a home in the data center: it did. The questions about long-term viability have been answered: it’s sticking around.
Since hyperconvergence is here to stay, and has gotten so big, the next set of questions has to be asked, starting with: what are the pros and cons of moving to hyperconverged infrastructure (HCI) in your environment?
Simplicity. This is one of the first things that come to mind with HCI. Imagine the traditional method of building out your datacenter. It involves buying a lot of disparate pieces of hardware and software, including:
- Management software for both the hardware and software
- Virtualization software, including a hypervisor
And that’s only the beginning of the list! All that stuff has to work together, communicate properly over the network, be backed up, etc. It’s headache-inducing.
HCI eliminates that huge pain point. HCI systems are fully integrated, with hardware, software and systems already tested, seamlessly stitched together and guaranteed to work as soon as they’re set up. Management nightmares vanish, since everything is created from the beginning to work together.
HCI is truly plug-n’-play, which is why many accurately refer to it as “data center-in-a-box.”
Scalability. We are living in the cloud age. Most businesses have experimented with cloud computing, and a majority have implemented some kind of cloud environment, even if it’s just for a small subset of workloads.
Scaling up can be a problem with mismatched hardware and storage. Different systems have different requirements, and older equipment may not work well with the newer hardware, whether it’s a CPU or storage array.
HCI has no such issues. Since it’s an integrated whole, scaling up is as simple as adding another node to the rack. Plug it in, turn it on and voila, more compute, more storage! No configuration hassles, no hardware compatibility checks. It’s almost like running a mainframe — it all works together as one system, and the more nodes you add, the more powerful it gets.
Here’s the thing, though: that increase in power and storage comes without any corresponding increase in management requirements. Since HCI functions as one system, managing a multitude of nodes is no harder than managing a single server.
Costs. HCI appliances typically use commodity hardware, instead of special-purpose components that ramp up the cost significantly.
In addition, you won’t have to hire additional IT staff to manage your HCI environment, since management is simple. The savings to your bottom line of less manpower could be significant.
Remember also that costs are more predictable. You know exactly how much it will cost to add a node, and how much additional capacity and power you’ll get out of it. It’s a pay-as-you-scale model, but those payments won’t surprise you. This is a hidden benefit that many don’t consider.
One throat to choke. Every administrator knows the pain of troubleshooting a data center problem: the server vendor blames the software vendor, the software vendor blames the storage vendor, the storage vendor blames the networking vendor, who blames the server vendor, and around it goes.
HCI has one vendor. One call when you have a problem. Someone who can’t blame someone else. That sound you’re hearing right now? It’s called “bliss,” and it comes from having a single point of contact for fixing issues that arise.
No solution is perfect, and there are some things to be aware of before taking the HCI plunge.
Vendor lock-in. This is the flip side of the “One throat to choke” benefit. When you use HCI, you are placing your bets on one vendor. That’s why it’s crucial to make sure that vendor has a track record in the space, and is well-regarded within the industry. A solid reputation can go a long way toward overcoming the concerns about lock-in.
Cost (again). This may seem odd, as it was also listed in the “pros” section. But if you’re not careful about your implementation and the vendor you choose, costs can spiral out of control. Some vendors charge a premium for their equipment and services, and cloud costs can also ramp up quickly.
The best way to manage your costs is to do a thorough cost/benefit analysis before starting. It’s not enough to consider CapEx (acquisition) costs; a good total cost of ownership (TCO) review includes OpEx (continuing operational) costs as well.
In other words, go into HCI with your eyes open. It’s not a silver bullet solution or panacea. But if you do your homework properly, in most cases you’ll find that the pluses far outweigh the minuses of an HCI implementation.