Is hyperconvergence software or hardware?
[dropcaps type=’square’ color=’#ffffff’ background_color=’#e04646′ border_color=”]T[/dropcaps]o make a long story short, it’s both. A hyperconverged platform will always have a software component and may sometimes have a hardware component as well. Vendors will support their choice to include or not include propriety hardware as the best and only correct option; but the truth is that the “correct” choice comes down to what you prefer.
The VSA (virtual storage appliance) that most hyperconverged offerings use to pool storage, or the kernel module in the case of something like VMware VSAN, are examples of the software that make a hyperconverged platform. On the other hand, some solutions like SimpliVity’s OmniStack do include custom hardware like the OmniStack Accelerator Card. The purpose of custom hardware is typically to increase performance and/or efficiency. In the specific case of the OmniStack Accelerator Card, the purpose is to offload data efficiency and integrity processes so that the system’s main CPUs don’t lose valuable cycles performing those operations and can continue serving virtual machine workloads.
Again, there’s not a right or wrong to whether or not hardware is included in an HCI platform. If special hardware is included, it dramatically limits your choice in regard to what equipment can be used to run the platform. But it likely increases stability, performance, and capacity on a node (all else being equal). The opposite view is that leveraging a VSA and no custom hardware opens up the solution to a wide variety of hardware possibilities.
While flexible, the downside of this approach is that it consumes resources from the hypervisor which would have served VM workloads in a traditional design. This can add up to a considerable amount of overhead. Which direction ends up being the best choice is dependent on myriad different variables and is unique to each environment.