One advantage of a hyper-converged architecture is the time to value. These systems are turnkey -- servers, storage,...
networking and software are all in one package -- so the time to create the first virtual machine is often measured in minutes. In a traditional storage deployment, the same process could take days.
But relying on the speed and simplicity of hyper-convergence also means accepting its shortcomings, one of which is lock-in with the provider. And the customer is not just locked into hardware -- the network and hypervisor are involved. It's possible to overcome some of these lock-in concerns, but it's at the risk of reducing simplicity and deployment speed.
Putting a hyper-converged, software-defined storage vendor to use
In large part, most hyper-converged platforms can also be considered software-defined storage that is designed to work with a specific hypervisor and bundled with network and server hardware. There are plenty of software-defined storage vendor options that are deployed within a hypervisor, but are not bundled with hardware. With these offerings, an organization can provide the hardware of its choosing or leverage its existing hardware. It can also mix and match hardware in the future.
Hyper-converged systems make deployment easy by prepackaging data center components -- but that often means IT shops can't use existing hardware.
The downside to choosing a software-defined storage vendor that only sells software is that the IT team has to assemble all the equipment, implement it, and install the hypervisor software and software-defined storage. There may also be special configuration work that needs to be done so the software-defined storage product can recognize the storage media internal to the cluster's nodes. None of this work is impossible and is likely within the skill set of an experienced IT professional. The biggest issue is the time it takes to perform these tasks, especially if something does not work as planned.
To overcome these challenges, it's easy to find a software-defined storage vendor that also provides hardware with its software for a turnkey approach. The result may be the best of both worlds: rapid initial deployment and flexible expansion in the future.
Breaking hypervisor lock-in
Even with a software-defined storage vendor, organizations will experience some lock-in. Most will only work with one type of hypervisor; if multiple hypervisors are supported, a separate cluster will be required for each one. At some level, an organization is locked into one or two hypervisors. There are products that will allow a seamless migration of virtual machines from one hypervisor to another or to the cloud. If these products are combined with software-defined storage, an organization can achieve a relatively high level of vendor freedom.
How free do you want to be?
Unfortunately, freedom within the data center typically comes at the cost of complexity. The more flexible an environment is, the less simple it is. At some point, one vendor will be your primary for a certain part of your data center, and you will experience some level of lock-in. There is a balance between lock-in and freedom that an organization needs to find so it can efficiently use its human IT resources.
Software-defined storage increasingly coupled with proprietary hardware
Characteristics of a software-defined platform
Software-defined storage basics: Definition, roadmap and use cases