The balkanization of virtualization will de-throne VMware, which doesn't mind a bit

ANALYSIS VMware is about to lose its status as the undoubted leader of the virtualization market after a 20-year reign – a result it doesn’t mind one bit, but which also signals an industry-wide end to significant innovation for traditional server virtualization. The end of VMware's reign will come slowly, then suddenly, between …

[Anonymous Coward]

just wait.

It will only take one stable full-parity competitor to erase VMware.

[Fred Daggy]

Re: just wait.

Doesn't even have to be full-parity. Just good enough at the features circa 80% want.

But, even as Broadcom liquidate VMware, they'll never get the "hearts and minds" ever again. Not that they care.

[Nate Amsden]

Re: just wait.

Realized late last year that many competitors still can't match fully what vSphere 4.x(last vmware product that I got excited about) could do sixteen years ago...seems clear most of them just stopped trying(perhaps until recently). Seems HPE VM Essentials is the only one trying to offer an alternative to VMFS for block based SAN storage platforms(using GFS2 and they do seem to be putting work into it I saw an announcement they can now do 2-node clusters). VM Essentials was pretty rough around the edges when I tried it earlier this year. Though I learned more recently, that verge.io supports fibre channel(not iSCSI) as well, though it's kind of a HCI layer on top(they may add functionality to allow customers to do without that added layer next year). Talked with Dell earlier this year they are going for this market as well(using a product they built or acquired many years ago originally targeted at edge solutions, it was called Native Edge but now it has a new name I forget), I think they support iSCSI now, though no idea when/if they may decide to support FC(they also have a "HCI" design I think even with iSCSI). Guessing none of them support VAAI-like array-based hardware storage acceleration(from vSphere 4.x).

Not that I need these days it but I looked a few times to see if there was any KVM-based alternative for vSphere's "Fault Tolerance" feature, again from 2010, and could not find any mention of it, other than some redhat post(s) from 2012-ish asking whether or not they should have that ability.

[Charlie Clark]

Re: just wait.

It's a market failure that customers let happen even as VM Ware was passed from hand to hand (including Dell). Not sure what the future will be, but we could soon be moving back to dedicated system setups with less general virtualisation.

[Doctor Syntax]

"Whatever the motivation for quitting VMware, users know they won’t find a better virtualization platform"

Define "better". Does your definition include "value for money" or "affordability"?

[EvaQ]

Good point, because a lot of VMware *users* (like system administrators) care less "value for money" or "affordability": they want the things they've been using and are experienced with.

Their bosses do care about cost.

Interesting how this will develop.

[retiredFool]

My experience has been in eda. There was a co called Mentor that really did have the best DRC/LVS, No one disputed it. Then Cadence started offering the following to its customers. We will give you our DRC/LVS for free. Cadence had the lock on P&R and layout. Oddly, the bean counters told engineering you will use cadence DRC/LVS and like it. Well, they did use it, but didn't like it. Because the bean counters stopped approving the Mentor purchases.

[l8gravely]

When you lose the bottom end, you lose the top end eventually

I strongly disagree with the theme of this article that "there will never be a better virtualization platform". When you lose the bottom end, or cede it, you're shooting yourself in the foot. It might not happen today, it might not happen for five years. But it will happen, because all the beginners and enthusiasts and small businesses where people learn will be exposed to KVM (or other technologies) and will happily stay there down the line. Especially as ProxMox and other vendors have all the incentive to improve their offerings to capture this large market. It might not be as lucrative, but it will still make them money.

IBM is in a lucrative niche, but it's also ceded the bottom end multiple times and look where it's ended up? Or look at DEC in the past? The bottom end is where people goto learn, experiment and build opinions on what is right and works. If you give that up... then eventually you will be sidelined.

If you're happy being a niche vendor, then that's good for you. But it isn't going to fly with the customers who only need basic virtualization and VMotion so they can run their envionment in a way that let's them move things around to replace old hardware, add in upgrades transparently, etc. That's it. 90% of the world only needs that.

And I also feel that virtualization will now explode in new directions, because the king voluntarily abdicated.

[Richard 12]

Re: When you lose the bottom end, you lose the top end eventually

Indeed. VMware is a dead product line walking.

It will receive no investment whatsoever, merely the absolute bare minimum of maintenance as Broadcom sweat the remaining customers, cranking up the fees for the remainder as the others leave.

Support will effectively have ended within three to five years when all the VMware staff have left along with their institutional knowledge. A couple of years later they'll take it out behind the woodshed.

It's a very old playbook, and one Broadcom know well.

Any customers who aren't already planning to leave are remiss in their duties. They won't go immediately, but nobody even vaguely sane sticks around for the second set of usurious price increases.

[Doctor Syntax]

Re: When you lose the bottom end, you lose the top end eventually

The staff running VMware at the customers' end should start learning alternative products.

[Energyx]

They went the way of Novell, and I saw it coming years ago. Lose focus on core products, get greedy and sell out. It's sad that some of the best technology and cultures in the industry end up gutted. From Brainshare to VMWorld, it was a similar decline.