Hyperconverged Infrastructure

Hybrid Cloud and Hyperconvergence Offer the Best of Both Worlds


Companies moving to cloud computing soon realize that they have to make another important decision: what type of infrastructure is best for supporting the new direction? And for many, they realize after a little study that a hyperconverged infrastructure, also known as HCI, is an easy choice.

This is especially true for businesses implementing a hybrid cloud strategy. Hybrid clouds offer the scalability and flexibility of the public cloud, along with the ability to keep data secure and the company’s data center under its ultimate control. It’s truly a best-of-both-worlds scenario.

There’s no getting around the fact, however, that increased complexity comes along with hybrid clouds. This is why HCI is such a good match. The promise of HCI is reduced complexity, since all your compute, storage, networking and the hypervisor are combined into one package that’s guaranteed to work out of the box. In way, the HCI appliance mirrors the cloud, in that resources are pooled and abstracted away from the underlying hardware, making the promise of self-service IT a reality.

With hybrid cloud, the public cloud platform becomes a component in your infrastructure, addressable locally. This means your applications can interoperate with the public cloud in the exact same way they would with any local system.

This means your applications can interoperate with the public cloud in the exact same way they would with any local system.

This has some big inherent advantages. For instance, consider backup and disaster recovery (DR). Since those public cloud resources become an extension of the local data center, they’re effectively on the LAN (not literally, but they're reachable as if they were). This means the networking nightmare for which DR is justifiably famous goes away, since an application that fails over to the cloud can conceivably keep its local IP address. Because of that, all the other applications, users and systems continue to communicate with the failed app as though it never moved. They don’t know whether or not the app is local or in the cloud, nor do they care: it just needs to work.

This represents the promise of true hybrid cloud: with simplified networking, an application with several virtual machines (VMs) can have them in the local system or in the cloud, with the same configuration. As with DR, the application doesn’t care where the VM resides. This is the fulfillment of the marketing adage “any app, anywhere” that was popular a few years ago.

Portability is drastically enhanced as well, since moving an application to the cloud just means live-migrating a VM between on-premises servers; remember that, from a networking perspective, the cloud (at least the part of it that matters to you) is on-premises.

Another plus when it comes to HCI and hybrid cloud is scalability. Since HCI nodes are purpose-built to work together seamlessly with no setup beyond plugging them in, you can ramp up your capacity in a snap. As your cloud needs -- or your on-premises needs, or both -- increase, your complexity need not increase. You’re still managing everything from a single pane of glass since you’re using HCI; there’s no need to orchestrate all the compute, storage and networking resources, hoping they’ll play nice in their sandbox.

It’s likely in the future that most organizations will opt for some sort of hybrid cloud solution, since it offers the best balance of in-house requirements with the ability to leverage the vast resources of the public cloud. And as we’ve seen, HCI provides the ideal platform for that solution in most cases.