Kroger has moved to SUSE and Azure Local

> Tottenham Hotspur Stadium has 20,000 network access points, 1,849 IPTV screens, and 519 CCTV screens. It hosts about 63,000 viewers

A decent part of that is probably important to be working on the day of a football match. They already own a suitable building, probably with decent backup power, so running their own servers seems both more reliable and cheaper than renting.

I would argue that absolute insanity is surrendering all control of infrastructure to cloud providers. Especially when you are using managed services that lock you in, making any plans of migrating way ridiculously painful.

It takes a CTO with cojones to actually own your infrastructure.

Nonsense. Renting servers is the easiest way to bankrupt your company. Look at all the capacity shortages going on right now. Price will continue to outpace bare metal and controls will get tighter.

Not to mention there are many scenarios where cloud connectivity is not an option.

Takes real balls to go with AWS huh?

> Obsession with maintaining some degree of physical control over infrastructure drives many organizations to absolute insanity.

Well... thank the current US administration for that one. The move they did with that ICC judge, cutting him off from anything digital, was a wakeup call for us Europeans that we can take nothing for granted any more.

Yeah it means nothing without details. They're a new HPE customer so they might be getting a discount the first year or something.

A friend of mine is a virtualization SME and from what I can tell you don't really save that much on the alternatives. Also, HPE sell VMware. HPE has a modular solution where you can pick the hypervisor you want to use, including VMware.

At my $dayjob we use Proxmox and it's good enough. I think Proxmox would be good enough for most orgs.

But my friend works with the tax agency and they definitely make use of vSAN, NSX, and all those nice features that puts VMware above the competition.

>The organization is saving “north of 85 percent” in licensing fees compared to when it relied on VMware, the executive said.

Nothing we can't google.

VMWare VCF is, let's say, $350/CPU core.

MVE is $600/socket, so at 32 cores, that's $18.75/core. At 64 cores, $9.38.

An 85% reduction in licensing/support fees is plausible.

100% was already possible, just not with 24/7 support.

And none of the quotes from that Pickering guy actually made any sense

To me it doesn't matter if it costs more or less to migrate. It's worth infinity to walk away from anyone who abuses you.

It's also worth infinity to switch from anything proprietary to open source. The claimed advantages of whatever proprietary product never really exist. You always get more value by paying developers instead of paying rent.

...And tolerating license audits. It boggles my mind that people are willing to tolerate some douchebags from MS or anyone else to come in to your office and turn over every rock and look through all your books, and you PAY them for this!

Yes, much needed to survive in Championship next year.

How much do they spend on players overalls?

Broadcom has absolutely zero interest in the long-term future of VMware. Their only goal is to squeeze as much money out of it as possible before it goes bust.

Let's say that VMware had a $10B revenue with a $9B operating cost. If 0.1% of their megacorp customers is responsible for 20% of that revenue while only being 1% of the support needs, then ditching the other 99.9% of customers reduces revenue to $2B while the operating cost can be reduced to $90M - increasing profit from $1B to $1.91B.

Those huge customers are quite locked in, so you can squeeze them for a couple of years before they leave. They have their own in-house support teams, so you can cut all L1/L2 support people. You're killing the product, so you can cut all developers except a handful to patch CVEs. The smaller customers who are leaving are doing some after a massive price hike, so you get a nice one-time renewal bonus while they desperately try to move to alternatives.

No need to do any shorting when you're generating massive profits for a couple of years. The plan when VMware is dead? Cut up its corpse in tiny parts, sell them off, buy another company, repeat the same strategy. As long as the total money they manage to extract from VMware is more than its acquisition cost, Broadcom has succeeded.

I didn't realize how addicted some orgs were to VMWare until Broadcom bought them and I witnessed zero people migrate to alternatives. It's been one of the single biggest shocks to me in my career. Watching them get squeezed openly and ruthlessly and just accepting it. I've only ever seen Microsoft pull that off before.

Note that execs are usually given stock grants in part to ensure they are long rather than short.

But you'd not be making money shorting Broadcom (AVGO); they're up on the year and YTD. They're an "AI stock" now, doing custom silicon for AI companies, and that's all that matters.

VMware is just going to be squeezed, rather than trying to acquire new customers. The entire corporate market is almost a rounding error compared to how big the datacenter boom is.