HCI? HC why?

If you're looking at changing the way you deliver your IT infrastructure with a more automated, simplified delivery mechanism, try looking at this new breed of HCI.

3 recover bricked pc
Thinkstock

The IT industry is full of new ideas. The constant cycle of technology means new ideas always come along that promise to fix big challenges. Some of these ideas interest me straight away, because they clearly address a common problem. Others take time, needing a change in the business or technology landscape to show their true value.

Hyperconverged infrastructure, commonly known as HCI, is one such approach to technology. According to Wikipedia it was defined in 2012 by Steve Chambers and Forrester Research to describe “a fully software-defined IT infrastructure that virtualizes the elements of conventional hardware-defined systems.” HCI includes, at a minimum, virtualized computing (a hypervisor), a virtualized SAN (software-defined storage), and virtualized networking (software-defined networking).  

Initial HCI solutions have been provided as a complete package of compute, networking, and storage, a “brick” if you will. These bricks are scalable: By adding additional “bricks” you can scale compute, networking and storage in a linear fashion. Need more compute? Buy another brick. More storage? Do the same. Pretty simple stuff … and simplicity is a big part of the attraction of HCI. Simple connectivity allows for quick and easy scaling, removing the complexity of all that cabling and networking.

For me, however, that simplicity has come at a cost. If I just want more storage, why do I need more compute? Or vice versa? That’s not the only negative I’ve seen with HCI, either. What about businesses that already have an IT investment in compute, storage, or networking? Why would they want to rip all that out and place it into one small package—which often comes with a not-inconsequential price tag?

That has been the crux of the problem in making HCI an attractive investment. For an HCI solution to encourage me to replace my current infrastructure investments, it needs to deliver something beyond simplification and squeezing all the components into a small box. It needs to bring new ways of doing things, new ways to automate and orchestrate the delivery of my technology. It needs to make my IT more efficient, effective, and agile. HCI initially has not offered that.

With all that said then, why write this article? There must be something of interest in HCI, right?

Recently I’ve noted a couple of technology shifts that have affected my view of HCI. One comes from the vendors themselves. The other is the development of software stacks better geared to exploit the HCI approach.

The software stack change is interesting. If you look at what VMware and Microsoft (especially via Azure Stack) are doing, you can start to see a clearer picture of how the HCI market can shape into something that can drive simplicity, efficiency, and value into our business.

The thrust of those software ecosystems is simplicity, automation, and integration with cloud, with the aim of helping organizations to transition to a more flexible and desirable hybrid infrastructure.

A key part of this development is to allow you to deliver a more public cloud-like infrastructure in your own organization’s datacenter. Look at Microsoft Azure Stack, for example. Its aim is to allow you to bring an Azure-like environment—with an automated and orchestrated service catalog—into your own data center.

VMware’s own automation and orchestration tools have (intriguingly) expanded with their AWS integration, and aims to do the same as Azure Stack. Providing automated, cloud-like delivery of services with tight integration to powerful cloud infrastructure into your organizations is an attractive proposition.

It is these powerful, automated, simplified, and scalable software stacks that have made me rethink my view of HCI.  I have gone from HCI hardware needing a use to these powerful software stacks needing a hardware platform that can match them for simplicity, flexibility, and ability to scale.

This new breed of HCI platforms continues to provide simplicity of deployment, but also provide greater flexibility in their scale. Vendors are now looking at separating the compute and storage elements of their platforms, allowing you to quickly and simply scale using the same idea of “bricks” of resources, but allowing you to scale compute and storage independently.

These two changes, in my opinion are making HCI a far more interesting option. It’s a marriage of automated and integrated software stacks with a hardware platform built to exploit these capabilities. That’s a more appealing choice if you’re looking to bring a new, more agile approach to deploying IT infrastructure and technology to support an organization’s goals.

Will HCI be the choice for everyone? No. There is still a price premium and a requirement to replace much of your infrastructure with a single vendor’s HCI solution (although several are now happier to integrate with existing components). But if you’re looking at changing the way you deliver your IT infrastructure with a more automated, simplified delivery mechanism, this new breed of HCI—alongside these integrated software stacks—is certainly worthy of consideration.

It has certainly changed my view. Where HCI in the past had been of little interest, now it’s something I include in my thinking when developing infrastructure. If you haven’t looked at HCI as a potential solution, maybe now is the time to look again and ask not just “HCI why?” but “HCI why not?”

This article is published as part of the IDG Contributor Network. Want to Join?