Whatever choice of hardware you make – on-premises, cloud, roll-your-own, converged or hyperconverged – don't expect to save any money.
So says Keith Townsend, a consultant, speaker and writer who El Reg saw in action in Sydney, Australia, at the local VMware user group conference.
At the event Townsend gave a talk titled “When to Select Hyper-Converged Solutions” during which he said “There is no such thing as saving money. You move it from one bucket to another you'll spend it on opex or capex.”
“It's cost neutral,” he said. “You'll spend the money sometime. The hyper-converged vendors know what you spend on hardware today and how far they can push.”
Sometimes hardware spend isn't the place to look for savings. “SAP HANA costs $20m to $70m to implement and $5m to $20m of that is hardware,” Townsend said. You could argue over the price of that hardware, but the real savings come from looking at other aspects of a project, like the savings in database licences to be had.
Townsend spend two years as a management consultant and says he crunched the numbers during that time to support his assertions.
The main value proposition of hyper-convergence, he therefore concluded, is the “one throat to choke” support offering that sees the systems vendor answerable to hardware and software issues. The “one throat” proposition even survives alliances between hyper-converged vendors and server-makers, so Townsend thinks the the SimpliVity/Cisco and Nutanix/Dell deals aren't going to dilute your throttling options.
Townsend said if you value “nerd knobs” - the ability to fine-tune systems – hyper-converged systems might not be for you. There's a tradeoff here of course: a hyper-converged system is sold as simple, in opposition to roll-your-own rigs in which you need people who are aware of innumerable variables that can tweak performance. Whether you'll miss the ability to make such tweaks is hard to predict – perhaps until you miss them and start choking that throat you've hired. ®