It’s not just about fault domains; it’s about supportability (who do I have to tell at the Change Control meeting if I want to do an upgrade) and compatibility – support matrices aren’t massively wide; you could run old hardware on new technology, but not get vendor support.

For HP, moving forward with their own HCI solution that allows them to wrap their software stack around it is the way to go and be relevant in the HCI space. That said, I’ve never once seen HP win a competitive deal against the likes of Nutanix or SimpliVity with their Lefthand VSA solution. That may change with the new systems they are providing, but I think it will be tough for them to gain traction against the other solutions that are far more mature and offer more value and specific integration across multiple hypervisors.

I think its foolish to claim HP has a lead marketshare in HCI. Giving away the lefthand stuff, especially at the sizing point that they have, cannibalizes the SMB space where HCI fits really well. Its like Dell claiming they ship more storage than everyone else because they count the disks inside servers, its a shell game to make marketing claims. The reality is, no one really looks at the VSA model as a hyperconverged system, and more to the point, that solution isn’t competitive.