What now for VMWare? What now for virtualisation?

[Quartz]

Massive price increases, apparently.

Some of the comments are quite alarming saying that there have been massive layoffs and many developers gone.

So, what are you server wranglers going to do if the price hikes stick?

 

[john_smith]

Hyper-V?

 

[paradigm]

It was clear from the beginning of the Broadcom takeover that the long goal was to end up with a portfolio of behemoth sized customers, the kinds of customers that are licensing that many products that they can leverage their purchasing power for discount.

No small customers = far less support required (large enterprise will have in-house expertise). They simply don’t want SME any more, so will price them out.

We have dumped VMware installs for our clients in favour of hyperconverged Hyper-V, be that in Server 2022 S2D form or Azure Stack HCI.

My previous employer were a middle to large sized VMware house, and they started to jump ship to Nutanix as soon as the Broadcom acquisition was announced, the writing was on the wall…

 

[BongoHunter]

We're keeping some Vmware for now - but anything that can be made to work well on KVM is moving to that.

 

[Throrik]

We're VCPP with a decent sized environment >500 but <1k hosts, lots of NSX, some vSAN and all of the Aria suite etc, so we're a bit up the creek really, as the timeframe for us is short, but we have to go with a three year commit so we're bent over. I suspect they'll be a large scale migration away from VMware, just whether they decide if they can do it prior to the cut off or not. We have a large Openstack solution which can run most things, and a lot of KVM too. I think our main concern is can we do what NSX does.

 

[ZedPowered]

Quoting up our renewal at the mo. Expecting an increase but also not wanting the upheaval of swapping out hypervisors for circa 30 hosts…

 

[LostCorpse]

Hyper-V
Proxmox
Xcp-NG
are all alternatives

Broadcom are cutting there throats effectively, i get what they are doing but i think its a bit silly.

 

[khrall]

Working for a reseller, it's certainly generated a lot of work for me!

People typically looking at Azure Stack HCI / Hyper-V, Nutanix, with a few looking at RedHat Openshift, as an alternative to the VMware platform.

The main issue with the FOSS based solutions (Proxmox, Xcp-NG), is the support. A lot of enterprise level customers can't run the risk of their workloads on software that don't offer 24/7 mission critical support. Also the existing investment in their backup & DR software/environments really does limit what they can realistically look at without boiling the ocean.

Edit: Forgot cloud, there are always people looking to move more to the public cloud too.

 

[Throrik]

People typically looking at Azure Stack HCI / Hyper-V, Nutanix, with a few looking at RedHat Openshift, as an alternative to the VMware platform.

How do you find people react when they realise Nutanix doesn't support SAN? My director nearly cried, as he saw it as the obvious mature solution, but the fact we'd have to spend millions we've already spent on block was the deal breaker. I think we'll look at OpenShift, but not sure we'd go with it over our current OpeStack deployment.

 

[khrall]

How do you find people react when they realise Nutanix doesn't support SAN? My director nearly cried, as he saw it as the obvious mature solution, but the fact we'd have to spend millions we've already spent on block was the deal breaker. I think we'll look at OpenShift, but not sure we'd go with it over our current OpeStack deployment.

Fortunately for a lot of people, they will be synchronised with a hardware refresh & the end of their 5 year VMware contract, or they will suck up the contract uplift until the hardware is due for replacement, but it's a large upheaval.

We've had a lot of customers that weren't lucky enough to lock in on EA's before the increases, and now they are stuck between a rock and a hard place. If you're on a "Evergreen" storage array, there are only a couple of routes to go to keep that investment.

 

[LostCorpse]

The main issue with the FOSS based solutions (Proxmox, Xcp-NG), is the support. A lot of enterprise level customers can't run the risk of their workloads on software that don't offer 24/7 mission critical support. Also the existing investment in their backup & DR software/environments really does limit what they can realistically look at without boiling the ocean.

just on this bit with XCP-NG there seems to be a company / enterprise option : https://vates.tech/blog/introducing-vates-virtualization-management-stack/?utm_source=xcp-ng-com
not sure if youve seen this or what your view are on it. maybe i miss understand the last bit.

i can see a lot of the smaller and mid size trying something else and generating revenue and interest for possibly the less well known.

Nutanix from what i can tell have a bad rep reading around and going to sting customers after year 1 or 2 based on feed back :/

 

[khrall]

just on this bit with XCP-NG there seems to be a company / enterprise option : https://vates.tech/blog/introducing-vates-virtualization-management-stack/?utm_source=xcp-ng-com
not sure if youve seen this or what your view are on it. maybe i miss understand the last bit.

i can see a lot of the smaller and mid size trying something else and generating revenue and interest for possibly the less well known.

Nutanix from what i can tell have a bad rep reading around and going to sting customers after year 1 or 2 based on feed back :/

Ultimately it will be up to the individual companies as to which route they will go. There is no one size fits all to replace VMware for everyone. Larger FTSE companies tend to be very risk adverse with their crown jewels.

I'm aware of some of the Nutanix reviews, a lot of those are based on mis-sizing the platforms or unexpected growth.

I'm not knocking FOSS as an option, for many it will be a consideration, but I know a lot of customers would not be inclined to run mission critical workloads on them today, but I can see elements of environments migrating over to them.

 

[LostCorpse]

appologies khrall that was in appropriate of me.
it is the individuals/companuies responsibility to assess and do home work on thee potenial business partners to see what option best suits them

​

 

[khrall]

appologies khrall that was in appropriate of me.
it is the individuals/companuies responsibility to assess and do home work on thee potenial business partners to see what option best suits them

​

Nothing to apologise for, it's healthy discussion.

It comes down to the usual IT response "It depends".

The VMware stack is a very capable product set and finding a cost effective solution as a replacement is no easy task without some level of compromise, be that features, function, support or compatibility, IMO.

 

[andshrew]

I've not had to contact them for a number of years, but Nutanix support was easily the best support experience I've had when dealing with a big tech company (eg. Microsoft, Vmware). The person answering the initial call was the same person that could help resolve the issue, and they actually understood the product they were supporting and not just following a script.

But Nutanix wasn't exactly a cheap alternative back when we switched... perhaps it is now by current Vmware standards...

They still have their community edition if you want to evaluate: https://www.nutanix.com/uk/products/community-edition

 

[img]

Still not had final costs yet even though vmware think we have. Seems there is minimum 12 core from what i read on internet though.

Might go back to cloudstack with kvm for some things though and then just kvm for hosts going foward but cannot reinstall 7k hosts.

I dont trust what says until we see terms as we were told 3 years but i hear from others you can get 1,3 or 5. M

Seems total mess and price could be amywhere from 2 to 10x more depending on setup. When you have based eveything on ram and its now cores its ot going to be good.

 

[covenantuk]

Broadcom have come in with the heavy upheaval stick - it's complete chaos - we were due to renew end January and still can't get a proper quote out of them and this is a common scenario apparently.

I can't see it ending well for them. Yes inertia will keep big enterprise with them initially, like most places we have a pool of experience and sunk cost in scripts etc. But bigger enterprises have hardware refresh budgets. We're due to replace hosts, so we'll probably renew with VMware until 2026, repurpose the old hosts to Hyper-V under a 3rd party hardware support contract and use the interval to skill up the team.

Seems total mess and price could be amywhere from 2 to 10x more depending on setup. When you have based eveything on ram and its now cores its ot going to be good.

RAM? Previous licensing was per CPU, up to 32 cores

 

[Throrik]

Broadcom have come in with the heavy upheaval stick - it's complete chaos - we were due to renew end January and still can't get a proper quote out of them and this is a common scenario apparently.

We've finally just got ours back, with proper push back on zero discounts on top of what we've managed, gone from 1.5m-ish to 16-18m. So VMware has officially been canned.

 

[covenantuk]

Tenfold increase? That's insane. Assume you went all in on cores/CPU if you had a RAM based agreement? How many CPU/cores does that cover?

 

[khrall]

Still not had final costs yet even though vmware think we have. Seems there is minimum 12 core from what i read on internet though.

I believe it's worse than that, it's a 16 core minimum PER CPU, not per host. A dual 8 core server would need 32 cores of licensing!

Seems total mess and price could be amywhere from 2 to 10x more depending on setup. When you have based eveything on ram and its now cores its ot going to be good.

I am aware of several people recieving a 10x increase, if you were an education customer, it's even higher as there does not appear to be an education discount anymore.