Cloud & Hyper-V Predictions for 2017
The end of the year is fast approaching and it’s time once again to do the sort of thing that I laugh at when Gartner and Forrester do it — make predictions for the coming year. In the past, I called it “Hyper-V Predictions,” but this year I’ve expanded this out to include the Cloud.
Say Goodbye to Traditional PC Lifecycle Management
Traditional IT tools, including Microsoft SCCM, Ghost Solution Suite, and KACE, often require considerable custom configurations by T3 technicians (an expensive and often elusive IT resource) to enable management of a hybrid onsite + remote workforce. In many cases, even with the best resources, organizations are finding that these on-premise tools simply cannot support remote endpoints consistently and reliably due to infrastructure limitations.
Reviewing My 2016 Predictions
I think I did pretty well with my 2015 predictions, so I was feeling pretty confident this time last year. But as it turns out, my crystal ball is cracked right down the middle. Let’s see how I did:
- Windows Server 2016 Licensing Woes Drag On: I said that the licensing of Windows Server 2016 (WS2016) was too confusing, and it clearly is based on the long conversations that I overhear from my sales colleagues, taking 20-30 minutes to explain cores versus CPUs or hyperthreads, the need to count pairs of cores, the need to buy a minimum of 16 cores, and doing this doesn’t cost more than WS2012 R2. But I predicted that Microsoft would remove the feature differences between the Standard and Datacenter editions. Microsoft did not, so I was wrong.
- Cloud Platform System (CPS) Will Become Interesting: Based on my conversations, no it did not. And if anything, the 2017 arrival of Azure Stack probably stopped any interest in CPS.
- Microsoft Azure Sales: I am seeing much greater interest in Azure in my local market. Classes that I teach are selling out in days. Every quarter we see stories about Azure business increasing year-over-year by 100 percent — that might be 100 percent of a small amount, but it is growth, so I’m going to take my wins where I can get them!
- OMS and EMS: The decision makers behind OMS decided to go the path of System Center 2012, and killed the bundle of solutions using licensing and pricing. I saw a sudden, hard stop in interest in OMS from the few customers I had that knew of it. The bundles that might be of interest are limited to large enterprises only, via licensing restrictions, so OMS is a non-starter for me and my customers. Interest in EMS continues to grow, mainly because of the security products. EMS received a boost this year with further improvements in Azure AD Premium and Azure Information Protection (previously Azure Rights Management) and the addition of Cloud App Security (in EMS E5). I hear anecdotally that EMS sales are actually pretty impressive. I think this prediction is a half hit, half miss.
- Hyper-Convergence Grows: Nutanix appears to make lots of noise — who knows if this is backed by profits or not! But for us Hyper-V types, the arrival of Storage Spaces Direct (S2D) marks a turning point. It’s still very early days for certified hardware availability, but the early interest is significant. It seems to me that HCI is here to stay and grow.
- Most of You Still Won’t Work With … PowerShell, RDMA, Containers, and Nano Server all are treated as the norm at big events such as Ignite, but my real-world experience tells me that most still have never heard of or used these features or products. Barely any hands popped up at a WS2016 launch when I asked if anyone had heard of RDMA, and I still encounter a complete lack of PowerShell knowledge with most IT pros — not even knowing how to run a script!
- Surface: Microsoft has broadened enterprise sales, numbers are still good, Surface Book issues have been solved for the most part, Surface Hub is selling faster than it can be made, and the Surface Studio was a huge marketing coup, possibly putting Microsoft ahead of Apple as the “cool company.” I also wondered if Microsoft would realize that any hope of Windows Phone sales lies in the enterprise market, and it appears that it has realized this, as evidenced by Win32 emulation coming next year.
I think I got 4.5 out of 7 with last year’s predictions. I won more than I lost, so that gives me the fuel to make more predictions!
Ignite “Azure Stack-Palooza”
I don’t have inside knowledge on this, but there’s a pretty good chance that Microsoft will embark on another big cycle of Windows Server previews at Ignite 2017. But as we saw earlier this year, Microsoft talked very little about Windows Server 2016 during day 1 of Ignite, preferring to talk about cloud solutions.
Azure Stack is due to launch in “mid-2017” on heavily tested and controlled hardware solutions sold by HPE, Lenovo, and Dell. My bet is that at Ignite in September, Microsoft will spend a lot of time talking about its unique offering, which is a common platform (Azure Resource Manager) that can run on Azure in Microsoft’s cloud, or Azure Stack in a customer’s private cloud or a hosting partner’s public cloud.
I am not sure what the sales will be like. There’s the probable negative effect of the hardware being sold in a heavily certified and controlled bundle, much like failover clusters before Windows Server 2008 – individual pieces go up in price because they are in a bundle. And we have no idea how Microsoft will license Azure Stack – I’ve heard rumors that I believe, but I cannot share those, and it will be quite different for most customers.
Microsoft will continue to recommend Nano for Hyper-V hosts and storage nodes and assume that this is what we are deploying as default. I completely get the goal of Nano – I would love a lightweight hypervisor management OS for my Hyper-V hosts. But I have openly questioned this recommendation, mainly because of issues with OEM hardware, particularly NICs. Early adopters are already seeing issues with WS2016, and as usual, it’s caused by NIC drivers, firmware, and/or pre-configured settings. Figuring out what’s wrong with a Nano Server will be a nightmare, so I obviously wouldn’t recommend this installation option.
And despite the introduction of the Nano Server Image Builder, the process of deploying Nano Server is still very reminiscent of Linux and MacGuyver – I could use one of Jeffrey Snover’s favorite terms to say that you’ll end up with a snowflake OS deployment solution if you adopt Nano Server.
Storage Spaces Will Still Be Unknown
Storage Spaces is a great solution, offering great performance, stability, and all at a lower cost than the SAN. But Storage Spaces has a big problem – few have ever heard of it, and of those that have, most think that it’s just Windows RAID from Windows Server 2000 – a curse on all things servers and storage.
The big issue with Storage Spaces is that most people who require clustered storage attend Microsoft events or tune into webinars. Microsoft subsidiaries stopped talking to customers about Windows Server technical features 5 years ago because it’s a run-rate business, and they have hard-to-hit sales/adoption targets on cloud services such as Office 365 and Azure. The only other source of information for the customer is the system integrator or hardware manufacturer. The cold reality is that sales people already know SAN, it helps them hit their revenue targets, so there is no motivation to learn about or sell the lower margin/revenue Storage Spaces solutions that they manufacture or can sell.
Despite all the promise of Storage Spaces Direct, I don’t see this changing. S2D adoption will grow in cost-sensitive companies that are rich in engineering knowledge, such as service providers, but SAN will continue to rule the roost.
Nature of Azure Adoption
Let me be quite clear: lift and shift, a process where customers migrate their virtual machines to the cloud, does not happen and will continue to not happen. If you get the chance, listen to Microsoft’s Jeffrey Snover talk about cloud transformation; he eloquently explains how customers must change how they design solutions to obtain the benefits of the cloud. Doing things the same old way that they did back in 1999 (one web server and one database server per service) just doesn’t work the way that you want in the cloud and doesn’t return any of the value of the cloud.
This cloud transformation can only effect new systems. News of big customers, such as Boeing earlier this year, will sound like a lift-and-shift is happening, but only when you get past the headlines and the first few paragraphs will you find that it’s for a new system or future applications.
What about those of us in the IaaS world? The leading solutions continue to be new online applications built on IaaS instead of PaaS, and hybrid solutions such as backup and disaster recovery. These are the top three things I see going from conversation into deployments, and I don’t think that will change in 2017. Microsoft is struggling to reach out to developers in the enterprise – the company’s not even trying from what I can see, so infrastructure will be the main path to cloud for established companies, whereas start-ups (which Microsoft is emphasizing) will embrace PaaS.
Surface Phone Will Be Awesome and Fail
Once again, I’m departing from the Hyper-V and cloud track to talk about something we Microsoft-heads are interested in. Even the family cat in my home knows that Microsoft Surface Phone is coming in 2017, and that Microsoft will finally do what it should have originally done with Windows Phone by focusing on the enterprise.
The hardware will look amazing, something we have come to expect from Surface since Surface Pro 3. The spec will be out of this world. Win32 emulation might even be enough to make a phone a viable alternative to most PCs (obviously not gaming PCs or workstations). The price might even work, and enterprises might be interested.
And the sales of Surface Phone will suck.
The consumerization of IT (which is really just an overhyped American corporation phenomenon and something that occurs in start-up businesses) will negatively impact Surface Phone. The lack of apps will be talked up. If these phones are meant to be PCs, then Microsoft needs these phones to be sold like PCs, and Microsoft has been very slow to grasp the role of the channel in hardware sales, hoping instead that some teenager in the corner AT&T store, that is motivated to sell iPhones because of cash-back rebates, will sell Surface Phones to enterprise employees. PCs are sold by Microsoft partners, and partners sell stock via the distribution channel. Surface Pro only started to make an impact in business because of the introduction of authorized distribution and reseller programs.
I do not see Microsoft realizing this, and the old habits of over-talking Cortana (crippled by The Curse of Zune) and sales via the telecoms companies will finally kill off Windows Mobile.