Some sales/marketing/channel type in Microsoft will get angry reading this. Good. I am an advocate of Microsoft tech, and I speak out when things are good, and I speak out when things are bad. Friends will criticise each other when one does something stupid. So don’t take criticism personally and get angry, sending off emails to moan about me. Trying to censor me won’t solve the problem. Hear the feedback. Fix the issue.
We’re still around many months away from the release of Windows Server 2016 (my guess: September, the week of Ignite 2016) but Microsoft has released the details of how licensing of WS2016 will be changing. Yes; changing; a lot.
In 2011, I predicted that the growth of cores per processor would trigger Microsoft to switching from per-socket licensing of Windows Server to per-core. Well, I was right. Wes Miller (@getwired) tweeted a link to a licensing FAQ on WS2016 – this paper confirms that WS2016 and System Center 2016 will be out in Q3 2016.
There are two significant changes:
- Switch to per-core licensing
- Standard and Datacenter editions are not the same anymore
Per-Core Licensing
The days when processors got more powerful by becoming faster are over. We are in a virtualized multi-threaded world where capacity is more important than horsepower – plus the laws of physics kicked in. Processors now grow by adding cores.
The largest processor that I’ve heard of from Intel (not claiming that it’s the largest ever) has 60 (SIXTY!) cores!!! Imaging you deploy a host with 2 of those Xeon Phi processors … you can license a huge amount of VMs with just 2 copies of WS2012 R2 Datacenter (no matter what virtualization you use). Microsoft is losing money in the upper end of the market because of the scale-out of core counts, so a change was needed.
I hoped that Microsoft would preserve the price for normal customers – it looks like they have, for many customers, but probably not all.
Note – this is per PHYSICAL CORE licensing, not virtual core, not logical processor, not hyperthread.
Yes, the language of this document is horrendous. The FAQ needs a FAQ.
It reads like you must purchase a minimum of 8 cores per physical proc, and then purchase incremental counts of 2 cores to meet your physical core count. The customer that is hurt most is the one with a small server, such as a micro-server – you must purchase a minimum of 16 cores.
One of the marketing lines on this is that on-premises licensing will align with cloud licensing – anyone deploying Windows Server in Azure or any other hosting company is used to the core model. A software assurance benefit was allegedly announced in October on the very noisy Azure blog (I can’t find it). You can move your Windows Server (with SA) license to the cloud, and deploy it with a blank VM minus the OS charge. I have no further details – it doesn’t appear on the benefits chart either. More details in Q1 2016.
CALs
The switch to core-focused licensing does not do away with CALs. You still need to buy CALs for privately owned licenses – we don’t need Windows Server CALs in hosting, e.g. Azure.
System Center
You’re switching to per-core licensing too.
Nano?
This is just an installation type and is not affected by licensing or editions.
Editions?
We know about the “core” editions of WS2016: Standard and Datacenter – more later in this post.
As for Azure Stack, Essentails, Storage Server, etc, we’re told to wait until Q1 2016 when someone somewhere in Redmond is going to have to eat some wormy crow. Why? Keep reading.
Standard is not the same as Datacenter
I found out about the licensing announcement after getting an email from Windows Server User Voice to tell me that my following feedback was rejected:
I knew that some stuff was probably going to end up in Datacenter edition only. Many of us gave feedback: “your solutions for reducing infrastructure costs make no sense if they are in Datacenter only because then your solution will be more expensive than the more mature and market-accepted original solution”.
The following are Datacenter Edition only:
- Storage Spaces Direct
- Storage Replica
- Shielded Virtual Machines
- Host Guardian Service
- Network Fabric
I don’t mind the cloud stuff being Datacenter only – that’s all for densely populated virtualization hosts that Datacenter should be used on. But it’s freaking stupid to put the storage stuff only in this SKU. Let’s imagine a 12 node S2D cluster. Each node has:
- 2 * 800 GB flash
- 8 * 8 TB SATA
That’s 65 TB of raw capacity per node. We have roughly 786 TB raw capacity in the cluster, and we’ll guestimate 314 TB of usable capacity. If each node costs $6155 then the licensing cost alone (forget RDMA network switches, NICs, servers, and flash/HDD) will be $73,860. Licensing for storage will be $73,860. Licensing. How much will that SAN cost you? Where was the cost benefit in going with commodity hardware there, may I ask?
This is almost as bad a cock-up as VMware charging for vRAM.
As for Storage Replica, I have a hard time believing that licensing 4 storage controllers for synch replication will cost more than licensing every host/application server for Storage Replica.
S2D is dead. Storage Replica is irrelevant. How are techs that are viewed with suspicion by customers going to gain any traction if they cannot compete with the incumbent? It’s a pity that some marketing bod can’t use Excel, because the storage team did what looks like an incredible engineering job.
If you agree that this decision was stupid then VOTE here.
Hay Aidan
Completely agree with the storage stuff. How Microsoft continually balls it up on the licensing front is beyond me.
I do think they have put themselves into a corner though with things like storage replica. I was thinking great with SR I could replicate volumes which hold SQL DBs therefore no need for expensive SQL Enterprise (per core) Always on AGs. What this licensing change means is that it is expensive either way to provide HA / DR with most Microsoft technologies.
Cheers
John.
Thanks for sharing this information!
if…or when this licensing change goes live it will make licensing more complicated.
With Server 2012 they pushed storages spaces – it was not perfect but with 2012R2 added features (tiering, etc) and it was a viable solution.
Now after reading a year about the new cool storage features (storage replica, etc) they are changing the licening model – this is a punch the face, as you mentioned above it makes no sense to go with the microsoft storage stack with this licensing model i think netapp & co will get some new customers….
May be that’s why Dell bought EMC?
[Disclosure: Microsoft employee]
I’m going to dissent here. Let’s first understand that Storage Spaces is Hyperconverged Infrastructure (HCI) which carries many benefits.
If you want HCI with VMware then you will not only have to buy all the prerequisite vSphere licenses but then you must purchase a VSAN license for each host as well. All-flash? That’s a second additive license for each host.
Storage Spaces Direct seeks to achieve the same thing. Just as many have found VSAN to be more economical and efficient than a CAPEX heavy storage array, I expect the same to he true from Storage Spaces Direct.
Charging extra for HCI in the hosts — providing a virtual storage controller — is already an industry norm and I’m sure many customers will find Datacenter Edition as the building block for Windows HCI an attractive and competitive solution.
You’ve drank the Gartner Kool-Aid and are trying to re-write the facts to suit your argument. Storage Spaces Direct is (a) an alternative to SAN and (b) a means to HCI. As an alternative to SAN it will fail, because (1) SAN will be more trusted, as it is against Storage Spaces today, and (2) SAN is cheaper than tens/hundreds of thousands of dollars of licensing. As for HCI, it’s a small percentage of the storage market. It’s the VDI of storage. It will not dominate because compute and storage do not grow at the same rate. And I sure won’t chuck $6000-$12,000 in licensing every time I need some more disk. The feedback (public and private) is pretty strong on this, and you’re pretty well out-voted.
Wow. How do you explain Nutanix then?
A minor player in a huge market, that’s really good at making a name for themselves. It exactly the same as happened with VDI, and Linux replacing the desktop before that. Gartner/etc talk about something, the press latches on and that’s “the thing”. I’m still waiting on my Linux VDI that I can log into from my rocket car.
Bandwagon purchasing by non technical CIOs who saw or read their marketing material.
Anyone who reads Aidan’s blog could build a better solution for a fraction of the price, and most likely way more bang for the buck.
Late reply on this I know, but I was researching something else, and read this article. Nutanix allows for asymmetric growth, you can put new nodes into the cluster that are heavy on RAM/CPU/Storage/whatever, and it will in theory load balance across.
That being said, I’ve always been leery of HCI solutions because of what Aidan pointed out. Storage and Compute grow at different rates, and a couple procs worth of VMWare Enterprise Plus or Windows Datacenter is a mighty expensive license every time I need to add a couple TB of space.
I’ve done a bit of research into Nutanix. The product has generated a bad reputation from what I hear, in terms of stability and performance. They are masters at marketing, though. Check out other HCI options if this is something you are serious about. Nutanix is just one brand.
Kevin as Aidan has mentioned most people wont be deploying HCI and I certainly wont – its a false economy.
I am currently running clustered storage spaces in 2012 R2. Having made this investment and ditching the old school SAN vendors it is a kick in the balls to know I will have to license for datacentre to get storage replication. Thanks Microsoft!
I suggested a compromise a couple weeks ago that asynch. rep be in standard and synch. rep be DC only.
Kevin, thank you very much for contributing to this debate. I have to side with Aiden here as I really can’t see compute and storage growing at the same rate. Now, I’m genuinely worried about pricing up a decent Scale Out File Server that previously might have only required 2 x nodes and just £1000 worth of software. Note to Aiden, I love Microsoft VDI and have built it largely as a result of reading your blog on Windows Server, Storage, RDMA, etc. 🙂
S2D limited to the DC SKU is a bad decision, period.
The $22,932 USD premium (for the minimum 4 node S2D cluster) is MUCH TOO HIGH.
There are plenty of great use cases where a 5-6 node S2D cluster running Standard SKU, with an small 5U physical footprint, would be an ideal offering, even if S2D was an add-on license per cluster or node.
As an Application developer, for medium size business target clients, this type of offering would be have been ideal for a high-availability deployment for our back-end business solution.
Now a 2 node cluster with 3 SAS JBOD enclosure is the only thing that we can offer.
(Have just had our first failure of that config at a customer site, with problems identified on both cluster nodes, and it has been hell trying to diagnose the failure source, all the while the client is trying to use our software/run their business, with no comparable hardware available)
Storage Replica as a feature is something more appropriate of a DC SKU (just as remote synchronization/replication feature is often an add-on for SANs)
Is Microsoft really afraid of loosing a sale? Does Microsoft feel the pain of consultants and vendors trying to sell into the enterprise space? Both answers are ‘No’. This is a push to say ‘just move to Azure and we’ll make it cheaper for you’.
Per core licensing – who went back to 2005 and decided that was the way to deal with declining sales? If Server 2012 R2 does everything asked for, and 2016 is a huge cost increase with negative features, why would folks move to it?
Keep in mind that anyone buying any new or subscribed volume licensing (even for 2012 R2) after GA of WS2016 will be paying per-core.
Thank you – did not know this. With server 2012 R2 EOL being in 2023, it may make sense to continue purchasing 2012 R2 as long as possible simply to get around the per core licensing (well DC licenses)
If you purchase VL, then you will be purchasing WS2016 after the release of WS2016 – there will be no “WS2012” purchase possible under VL after the release of WS2016 – you simply opt to “downgrade”. So the rule will continue to apply to you.
I mean more of the purchase 2012 R2 VL while you can (before 2016 release) so that you can avoid the per core licensing!
Also of note, some of us are using Service Provider Licensing Agreements (SPLA), and can only purchase the current SKU. That means for us that the change requires a number of immediate decisions, even if we choose to use a well-known prior release.
The big flaw in the Microsoft reasoning is the fact that the amount of VMs does not scale 1:1 with the number of cores. More cores per socket does often mean slower cores for energy efficiency as the good old Ghz race was raising power cost exponentially instead of linear.
VMs vs Cores is not a linear rising line. Microsoft is charging money on processing power that is simply non existent.
Firstly:
> Trying to censor me won’t solve the problem. Hear the feedback. Fix the issue.
This will get you banned for life in some forums. And: People come by and call you names for that, with the applause of the moderators. Thats what the Microsoft-community is like as of today.
“If you don’t like the product, don’t buy it. And if you don’t buy it, you’re not a customer, and therefore not important.” – Thats the reasoning.
And that’s a fantastic way to minimize the number of customers. I would suggest a warning to stockholders before adopting that sales & marketing technique.
Microsoft’s documentation says that the 16 core licenses of 2016 standard or datacenter will be the same price as the two proc 2012 R2 version. So it doesn’t hurt the smaller guy it seems. Please correct me if I’m wrong.
What smaller guy? The guy who’s running his <150 user company on a pair of servers with clustered storage spaces? Sure! Except he doesn't exist. All of his stuff is in Office365/Google Apps.
Or the mom and pop shop that's using a micro server as a fancy NAS?
Actually, you are very wrong. Every CiB we have sold has been to a small-mid company.