Check out those whales. We’ve been seeing a lot of them right off the beaches in Half Moon Bay. It’s either a wonderful gift from nature of a sign of incipient ecosystem collapse. ¯\_(ツ)_/¯
Last week was VMworld and I trekked there for the yearly infrastructure homecoming and bacchanal. I laid off the steak but I did accept some Moscow Mules from vendors of various sizes.
The first question that everybody asks is: Will Dell leave VMware alone? to which the only answer we have so far is the only one that Michael Dell and Pat Gelsinger are willing to give, which is, Yes, Dell will leave its Golden Goose alone. The implication, of course, in all acquisition promises, is the continued existence of Golden Eggs.
But the real question people have is about the Golden Eggs. Does VMware still have them? Kurt Marko, writing for Diginomica, seems to think no. IT diehards seek shelter at VMworld as future rushes past. Kurt has been around long enough not to chase the latest shiny object, but the crux of his argument seems to revolve around two different worldviews, and indeed two different surveys.
VMware’s analysis showed that adoption of public cloud services was still small, amounting to 15% of total workloads this year, doubling to 30% in five years. … 451 Research released survey data from 32,000 senior IT professionals showing that “41% of all enterprise workloads are currently running in some type of public or private cloud. By mid-2018, that number is expected to rise to 60%…”
I was trained as a scientist to look at data. You would choose very different personal and corporate strategies depending on which number captured the world as it currently is. No wonder each side thinks the other are either frivolous hipsters or stodgy old ostrich-heads.
Normally I think that a lot of confusion around cloud comes from not mentioning years — are you talking 2026 or next quarter? And I’ve got a quota to make next quarter so stop inventing crazy stories about ten years from now.
But in this case, both numbers, the 15% and the 41%, both say they are about the present day, 2016. So we have to ask, we have two samples, but of what populations?
I suspect VMware’s number includes the long tail of smaller shops and non-tech businesses, the kind of places that would never be part of a 451 survey. But again, VMware didn’t reveal how it got those keynote numbers either, so ¯\_(ツ)_/¯.
I do think that apps and workloads are the right way to look at this. If you’re making greenfield apps, you are making them with Microsoft, who is pushing you to Azure, or OSS/Linux, which is leading you to AWS. But the legacy apps? That’s VMware.
So when I do my imaginary math on the back of my imaginary envelope, and I know there are still Windows NT apps and Windows 2003 machines running businesses — and neither of those have been supported for a while, and I know how long even health organizations hold on to their apps (as long as possible), so my imaginary math comes out closer to VMware’s than 451’s. I just don’t see how 41% of all the workloads in the world are in a cloud, unless we’re just counting vSphere as a cloud, and I hope we’re not doing that. I think we’re all going to have to transform, but I don’t think we’re all butterflies yet.
So if the event horizon is indeed farther away, is VMware prepping its engines to escape the singularity and make it back to known space? Based on what we saw at the show, I’m guardedly optimistic. (Disclaimer: I’m still long VMW.)
VMware had to do three things:
- Signal that they see the real world without rose-colored glasses. The talk of cross-cloud and multi-cloud this year was that signal. It’s a marked change from “One Cloud” of last year.Cloud Foundation is VMware’s signaling mechanism. It seems like more than just a bundle, and I’d expect other cloud providers to play along. This is the future of your current apps that live in VMs – a few clicks has vSphere + add-on services live in many clouds.
- Demonstrate that they’re going to be add value to that world. NSX is the key here. I interviewed NSX Boss Rajiv Ramaswami on theCUBE, and he pointed out that NSX use cases lay out:
- 40% security (micro-segmentation between apps)
- 40% automation (The Trend Formerly Known as SDN)
- 20% BCDR (cross-DC connectivity)
Now this does show NSX’s weaknesses: that the grand cross-cloud future isn’t here yet (it’s just the 20%), and also shows that NSX is a bit of a jack-of-all trades, which could make it harder to explain and to sell. But there’s a lot there from a Foundation standpoint to like.
VSAN is touted as part of the future, and it is from a revenue slice and from the future death-of-the-SAN, but I don’t see it as potentially cloud-transformative as NSX.
- Make noises for the container folks. Kit Colbert has good timing. The same week the Docker ecosystem is talking about a fork because of instability and non-production-readiness, Kit is talking about stability and production-readiness. That’s all he had to do for now. Container dollars are still pennies compared to cloud dollars right now.
So overall I think VMware can continue to be relevant if they can continue to be relevant and make friends in the clouds. Somebody has to be the grown-ups here.