Hyper-V Server 2022

Steel Contributor

Anyone know whether there will be a Hyper-V Server 2022? i.e. the free version which is just for running VMs and has no GUI?


I've seen mentions on forums that this SKU is being dropped, but not found anything official.



278 Replies
@ChrisAtMaf Using Server Core with the Hyper-V role requires an additional Windows Server standard license.

There may be an exception to needing an extra license if the Server Core with Hyper-V role is being used "solely" for host the licensed server, but then you lose a lot of flexibility with hosting an extra VM alongside, without getting into complexities of Nested virtualisation. You could also put your server on bare metal and then run other VMs off that, but we are trying to get away from bare metal.

The old Hyper-V Server was great because it just eliminated all these licensing problems and considerations and you could just go ahead and do it.

From a technical standpoint. Server Core with Hyper-V is basically the same thing. It is all about licensing. which is always complicated enough with Microsoft as it is.
That is correct. We start with Windows Server server core composition... and since it's only for a virtualization host we remove unsupported Roles / Features. Both Microsoft Hyper-V Server and Azure Stack HCI are a reduced set of server core. For Azure Stack HCI there is fewer features removed, such as Storage Spaces Direct and SDN. Then we add some Azure integration layers on top, like Arc. So they are nearly equivalent.
@Elden Christensen

RE: Feature removal of Azure integration features

So, the core question you should be asking yourself here is "Do we want Enthusiast users who don't mind using pre-release software for free be able to get their toes wet with Azure, or are we going to make them switch to a 60-day trial and then pay for it in order to do that?"

Keep in mind:
1. Trying the Azure features will, at minimum, mean that an Azure account will get created. That is a foot in the door.
2. By using those Azure features, they will become more familiar with them and go deeper into the ecosystem.
3. As they go deeper into the ecosystem and start building things of higher importance/value, then the prospect of installing a supported/paid version of Azure HCI becomes better.
4. When the Azure credits get used up or expire, this could very well result in a paying Azure customer to keep using those paid Azure features, even if it's on an unsupported system (or they upgrade to a supported one).

No problems missing out on non-Azure features like Storage Spaces Direct and SDN on a free tier. They weren't even in Hyper-V Server 2019. If anyone wants to use those, that is a good use case to use a trial/paid version of Azure HCI.

It's your call. As I said before, it doesn't even affect me anymore because I have already moved on. I am just looking out for your best interest (to do a good job for Microsoft) if you were wanting Azure/Azure HCI to be as successful as can be. I think that you would want it to be easy to get hooked into Azure by making less barriers.
So, it just occurred to me that the business model for Azure Stack HCI is all wrong.

The most straightforward way is to shift the payment from being centered on the OS, itself over to what actually connects to Azure.

i.e. Instead of paying $10/core/month for the OS, you pay $10/core/month for each core *LINKED TO AZURE* (each core gets a 60-day free trial).

The Azure Stack HCI OS would still have basic functionality without the connection to OS roughly equivalent to Hyper-V Server 2019.

When you do link your HCI to Azure, that will unlock additional functionality, such as:
* Not be forced to install pre-release software on Patch Tuesdays
* Storage Spaces Direct
* The actual functionality which Azure offers

If that link to a paid version of Azure is not maintained, then those extra features become deactivated.



I think the whole Lab/Home Use thing is a bit of a red herring. We have always had the ability to install the current version of a Windows Server Core (for any service) in an evaluation/lab/non-production environment. There is nothing about this move that prevents you from running Server 2022 Core w/HV Role, in a lab or at home to test out or experiment with features and management. I don't see that ever changing, in fact Server Eval. has gotten more open and better over the years.


The problem I have is the impact on licensing a hypervisor, either by updating the VM licensing before it would be necessary, or via an Azure subscription on more expensive hardware. This is the largest impact to SMB. 

> What if there was a mode that had significantly reduced functionality but allowed basic VM hosting? Such as maybe removing enterprise grade / production critical features... like clustering / HA? VM mobility? DR capabilities?

Yes. That's what significant portion of people here would love to see. Separate product "Azure Stack Lite" or installation option, without the features you mention (although keeping VM migration would be sweet), and provide that product on the same licensing terms as Hyper-V Server 2019.

As I wrote before, we have currently handful of Hyper-V Servers both in lab and production, that VMs (licensed in various ways) get moved onto and out as needed. And there's currently not a good and proper way forward with this scenario without incurring additional licensing costs.
Yes, very fair point. Using Windows Server eval with the Hyper-V role and server core install is going to be nearly identical to Hyper-V Server. Which gives 6-months free for home use.

We seem to be flailing around rather wildly on this thread, especially with all of the traffic today.  Correct me if I'm wrong, but when I try to distill the thread down to its core components, what I basically see is:


1. Someone asked about whether a free Hyper-V Standalone 2022 server would be a thing.

2. The only Microsoft person in the thread, @Elden Christensen , pointed the OP at Azure Stack HCI as the answer to that question, which implies that ASHCI is replacing Hyper-V Standalone.

3. Lots of people - myself included - posted reasons we didn't like that.

4. Microsoft (Elden) basically said, "If ASHCI doesn't work for you, we'd like feedback on how to make it better."

5. Lots of people replied, but our replies have boiled down to, "ASHCI doesn't work for us because it isn't free!"  

6. Microsoft hasn't responded to that.  In my opinion, that is because they've already chosen how to proceed, and are trying to get us to move there as well.


In this thread here has been lots of discussion about aspects of hypervisors, and reasons we all need or want the legacy Hyper-V Standalone product, but what it seems to boil down to is this: 


The reason we all like Hyper-V Standalone is because it is free.  Period.

The reason Microsoft is moving to ASHCI is because Hyper-V Standalone was free.  Period.


I do not work for Microsoft, but if you look back on Microsoft's responses in this thread, they have done everything by way of explaining aspects of ASHCI, pointing people to resources about ASHCI, and answering questions about it.  But at no time has Microsoft said, or implied that, as a result of this thread, they're going to "bring back" (or "bring forward"), Hyper-V Standalone server.


To me, at least, that says that they're not going to.  It's dead.  Period.


As an aside, I was intrigued by the reference to the Partner Program in this thread.  I'm sure many of us in this thread are members of that program too, just as we are users of Hyper-V.  I hadn't yet heard about the changes to the program, but it turns out that Microsoft is massively reworking it.  The end result of that rework is that far fewer companies will be able to qualify to be what they are now calling "Solutions Partners" - and it will be far more difficult to keep that qualification.   In essence, right now there are hundreds of thousands of "Microsoft Partners" who - like me - qualify simply because it was easy to do so.  We could sell stuff, or we could certify.  We didn't have to do all of it.  Under the new program, we have to do a lot, and a lot more than we used to.  So, like most of those smaller partners, I'm going to lose my status when the new program takes effect.


Why, you might ask, is Microsoft doing this? Only they can answer, but I think part of the reason is that the current partner program is watered-down:  Microsoft spends a lot of time/money/energy on hundreds of thousands of little partners that do not make Microsoft any real money.  So, Microsoft has chosen to change things:  Now, there will be far fewer qualified partners, but they will get better attention and better benefits. 


To me, the same holds true for Hyper-V server.  Tons of people use it.  Using it, they have expectations surrounding it.  Meeting those expectations costs Microsoft money.  But the reality here is:  It is not *making* any money.  We can all say things like, "Oh, well, Hyper-V led me to Microsoft, and led my customers there as well!" but I suspect that all of the "us"'es together with our customers still don't cost-justify Microsoft releasing and having to deal with a Hyper-V Standalone server going forward. 


Microsoft and other companies change things all the time - heck, I'm still grumpy that Samsung dropped MicroSD card support from their Galaxy S line!  It makes no sense to me:  They *make* microSD cards!  But it clearly makes sense to them, and no amount of complaining is going to change it.


So, what's my point?  My point is:  Microsoft is trying to answer the questions here, but if you read back through the thread (although some posts have been clearly moderated away), it is clear that Microsoft has killed Hyper-V Standalone server.  They'll listen to feedback, improve ASHCI where they can, and do their best, but they've decided that Hyper-V Standalone isn't cost-justified, and I can't say that I blame them.  But whether I can or not, Hyper-V Standalone is not coming back, any more than Gold Partnership is coming back.  It's dead.  It's done. 


If I had one complaint, or suggestion, for you @Elden Christensen and Microsoft, it would be this:


Just say that.  Clearly.  Here.  On this thread.  Tell us: "Hyper-V Standalone is gone, and it's not coming back."  I can't speak for Microsoft, and I know people here won't like it, but we're flailing around because some of us think we can somehow "convince" Microsoft to give us our free product back again.  But I'm guessing that's not going to happen, is it?  Some confirmation would be helpful in bringing us closure.


As for "What to do now," to me it's pretty simple:  For myself, I will use Hyper-V under Windows 11 to run my Windows (and other) servers in my own home and labs, just as I do now.  Windows 11 is not free, but it's pretty cheap, and it's a buy-one-and-done, and I can run Windows servers under Hyper-V under Windows 11.  What's retail for that?  $250?  Most of us get it for free on our existing computers.


It was fun having an actual "server" with servers on it - I enjoyed Hyper-V Standalone too - but those days are gone.  If I still want an actual headless server instead of a server running Windows 11 as the host, I could always go to what I use at my dayjob:  XCP-NG works really well, Windows servers run happily on it, and you can migrate and do all the other stuff you're used to doing on Hyper-V Standalone.  And XCP-NG and it's control plane "Xen Orchestra" are all free and open-source, forever.


The older I get the harder it is to change... but the older I get the harder it is to get upset about change.  None of us in this thread like it, but Microsoft is doing what they perceive to be best for Microsoft and their shareholders (of which, I'd guess, most of us are also one!).  They're not a monument to justice.  They're a corporation.  They will survive, and we will survive, and even though things are bumpy I hope they will get better.


Meanwhile all we can do is make our own business decisions.  For me and my team, despite Hyper-V Standalone going away, despite the Partner Program going away (or going much farther out of reach), we've determined that we're going to roll with it, and do the best we can with the cards we're dealt.


Which, really, is all any of us can do.


By all means - if I'm wrong, I hope Microsoft will correct me here.  I hope they'll bring forward Hyper-V Standalone.  I hope they'll keep letting me be a Partner.  But I'm not basing my actions on it, because Microsoft is big, and we are (comparatively) small, and that's just the way it is.


Good luck to us all as things move forward.   Consider Win11 Hyper-V if you can, or XCP-NG if you can't.

@Elden Christensen 


While a valid scenario, the home lab basis for this thread's resurrection is becoming a distraction. There are other paid scenarios left in the cold such as small Microsoft Partners (small shops or consultants like myself) on the paid-for Action Pack pack subscription that have been left out in the cold by this change. I don't mean "paid" for the hypervisor, just recognition that from a contractual perspective, we're not running off free, unsupported trials.


There is a very real cost impact here that you can probably quantify better if you can get hold of the sales figures for Action Pack subscriptions - I certainly can't guess it but it would be more than a few.


No Hyper-V Server equivalent (cost-wise and technically as has already been well-covered months ago regarding single node) directly affects our ability to run a Microsoft hypervisor, which given we are Microsoft Partners, is something we very much want to do so we can actually advocate for such products. Advocacy is a very big part of what we do.


Hyper-V Server didn't get to where it did - making HCI a decade-and-a-bit later than much more viable as an option - by accident. It was (and is) a good product with a lot of advocates pushing back against the "gold standard" vendor based on direct use-based experience for over a decade. That advocacy will inevitably shift under the current scheme.


I can only speak for myself and a few other Partners in this boat, but to be crystal clear: we are not interested in anything DR or HA for these small but persisted (i.e. not a trial - that's also a non-starter as we have Action Pack accessibility) environments. Support (or lack of it) is not a profound issue, either, though for business customers (small or otherwise) is most certainly is.


There are many other scenarios - all well-covered earlier - for which HCI is a non-starter in its monthly subscription format.


While as a Partner I'm content to remain on Hyper-V 2019 for the time being, with no equivalently-costed option remaining, we will have to move if that remains as-is and it won't be to any vendor's "pay-per-view" model.


I feel this was mentioned months back, but a better commercial starting point would be parity with Hyper-V Server with the cloud based services disabled (may be a technical impossibility now but I'm simply ruminating). Once a subscription is detected (directly; via something like VAMT which already handles offline Windows; etc.), that should be the trigger for feature activation as you've already hooked the customer on the OpEX business model everyone so badly craves these days.


So, while I've focused on small Partners here, please do not get so distracted to the point of exclusion by home-use scenarios such that potential solutions for small business and Partners alike are not identified through not being visible to you.




Good write-up.

I feel the same way about a final position statement.

Put it in here so it's a tangible reference point and let's let this thread rest in peace (if that's where things are at.)

@LainRobertson thanks, and I must say I agree 100% with your points as well.  Very well put. 


Hoping for the best for us all - Microsoft included!


best response confirmed by Elden Christensen (Microsoft)

Hyper-V in the 2022 Wave

Since its introduction over a decade ago in Windows Server 2008, Hyper-V technology has been, and continues to be, the foundation of Microsoft’s hypervisor platform. Hyper-V is a strategic technology for Microsoft. Microsoft continues to invest heavily in Hyper-V for a variety of scenarios such as virtualization, security, containers, gaming, and more. Hyper-V is used in Azure, Azure Stack HCI, Windows Server, Windows Client, and Xbox among others.


With the release of Windows Server 2022, Hyper-V is included as an in-box role in Windows Server 2022 Datacenter, Standard, and Essentials editions just as it has with previous releases for well over a decade.


We continue to innovate on-premises virtualization with a new product released in December 2020 called Azure Stack HCI.  Azure Stack HCI is a purpose-built virtualization host with deep hybrid integration.  It takes the Hyper-V host you know and love today and combines it with the power of Azure.  It takes those on their first steps in the journey to the cloud, empowers those who live in a hybrid world, yet is still familiar for those existing Hyper-V administrators in what you know and love.  The same tools, processes, and skillset can manage an Azure Stack HCI solution.


Azure Stack HCI is Microsoft’s premier hypervisor offering for running virtual machines on-premises.  We recommend that all customers using Hyper-V today evaluate Azure Stack HCI.  We have an exciting roadmap of features and innovation coming!


With the arrival of Azure Stack HCI, one change is that the free ‘Microsoft Hyper-V Server’ SKU which is a hypervisor host available for download is being discontinued starting with the 2022 version. Microsoft Hyper-V Server 2019 will continue to be supported under its lifecycle policy until January 2029, see this link for additional information:


For customers looking to do test or evaluation, Azure Stack HCI includes a 60-day free trial and can be downloaded here:  Azure Stack HCI is billed through a subscription model which can be cancelled at any time.  


Microsoft remains committed to meeting customers where they are and delivering innovation for on-premises virtualization and bringing unique capabilities like no other can combined with the power of Azure.  We are announcing that there will be no Microsoft Hyper-V Server 2022 version of the free download, giving customers several years to evaluate, give feedback, and take their journey to Azure Stack HCI.


Thank you,
Elden Christensen
Principal PM Manager
Windows Server Development Team


@Elden Christensen I'll risk a final bit of noise to say Thank You for making this clear.  We're all doing the best we can - we just need to know, clearly, what the situation is.  And now we do.  Thank you for that.


Thank you also for answering all the other questions.  For myself, I'll be looking at Server 2022, and ASHCI, and watching with great interest as Microsoft rolls out new products and features.



Didn't see any impact from all our use-cases for various things needed or issues with the plan with HCL - at all. Bummer, 2022 looks like the year of "pound-sand/kick-rocks" from MS on pretty much every front it appears.


So long and thanks for all the fish

Ok, Thank for this concret answer.
I'm student, and I'll never paid 10$/month/core (versus free).
I use replication and "hot swap" between machine (I'm french, I don't know the english word).

Fo me, the support of hv 2019 is for janury, 2024, because I'm not professional, so I don't have access to the premium support up to 2029.

It's the end of route for MS and I, soon...

Thank for all the versions and free licencing you build up to this day.
Technically, 2024 is only the end of the first phase of support known as "mainstream support". It will then switch over to "extended support" which ends in 2029.

During extended support, you will continue to receive security updates but no new product features.

It's not likely you'll need until 2029 to switch to another product but I thought it would still be good to explain the different support stages so you better understand your options.

@Elden Christensen wrote:

Note:  We also heard the feedback, which is important for your scenario to support 1-node... that's coming soon. you need full functionality?  What if there was a mode that had significantly reduced functionality but allowed basic VM hosting?  Such as maybe removing enterprise grade / production critical features... like clustering / HA?  VM mobility?  DR capabilities? 

Regarding what our needs would be, a 1-node scenario, plus a discount or grant for education / nonprofit usage so we can deploy a few without swallowing up our entire Azure grant in one bite. If you're considering a free mode with more limited functionality which can run in 1-node for branch offices, that sounds fair - perhaps with similar functionality levels to Hyper-V Server. It seems fair to remove failover clustering and high availability. It would be helpful to leave basic replication and maybe shared-nothing live migration (not failover cluster live migration) available in the base product, which would retain some semblance of feature parity with Hyper-V Server.


In terms of using Server Core with Hyper-V role, in reply to @SGGGG I think you might need to check the license for 'free' Hyper-V Server. It contains exactly the same limitation as Windows Server Core allowing it to 'only' be used to host licensed VMs - 'The instance of the server software running in the physical operating system environment may be used only to provide hardware virtualization services, and/or run software to manage and service operating system environments on the
licensed server' - so there's no licensing difference running Windows Server Core with Hyper-V (apart from the fact that your Windows Server license also allows you to run 2 (Standard) or unlimited (Datacenter) VMs as part of your license) when compared with running the 'free' Hyper-V Server. As @imschmidt mentioned the main benefit of using 'free' Hyper-V Server when you have Windows Server licenses is that you can make use of Hyper-V Server 2019 features regardless of the Windows Server license version you purchased.


For me though this is about Microsoft providing their 'next gen' hypervisor (where they will be focusing new features and support) at an affordable cost for enthusiast, SMB, education and nonprofit customers. As others have said, providing this at free/low cost allows these areas to remain in the Microsoft ecosystem. Without that product I think Hyper-V adoption within those markets will suffer a permanent and possibly irreversible decline.


There are other situations where the license matters, apart from the version - e.g. in a test environment where VMs are short-lived and being run on trial licenses, where non-Windows VMs (e.g. Linux) are in use, or in some VDI scenarios. In all of those cases the workloads do not require licensing, so Hyper-V Server is ideal for this. Server 2022 core with the Hyper-V role will of course work from a technical perspective, but there are licensing costs.

Different users are obviously going to have varying requirements, but from my perspective I want a basic hypervisor to run test VMs on (with a perpetual license - a 60-day trial is of no use) - advanced features such as DR and clustering are not required.

I really don't think the preview channel is appropriate - hypervisors need to be stable, and unless actually testing new hypervisor features use of the preview channel sounds like a very bad idea. Ideally it should be on the same update channel as Azure Stack HCI and Server 2022 (I notice that these two both share a category in WSUS now).

I also want something that will function as a standalone install - i.e. not requiring enrollment into an Azure tenant.
I have a very small shop. I have 1 server (Quad-core Intel Xeon) that I use to run a file server, DHCP/DNS server, and a few other special-purpose Linux servers for backups and such.

I do not know your anticipated target audience, but I am not going to pay $40/month for that. In less than a year, it would be more cost-effective for me to buy a Server 2022 Standard license, but I would be buying a lot more than I need.

I'm in the process of moving my 2016 Server Standard to the free 2019 Hyper-V Server, but I would honestly pay something like $99 (one and done. Please don't bill me monthly) for just the newest Hyper-V.

Just a thought since you do not seem to have a use case for users like me.