Posey's Tips & Tricks
How Will Windows Server 2019 Impact Hyper-V Architecture?
Microsoft's forthcoming Windows Server product looks to be jam-packed with new features. Here are some of the standouts, specific to Hyper-V.
As has been the case with every Windows Server release, the forthcoming Windows Server 2019 is jam-packed with new features.
I don't want to waste your time rehashing all of those new features; just check out some of Redmond's extensive coverage of the upcoming release here, here, here and here for a look at what's coming. Even so, there are a couple of things about the new release that caught my attention, especially with regard to the future of Hyper-V.
VM Shielding for Linux
When it comes to Hyper-V in Windows Server 2019, the feature that has probably gotten the most attention is virtual machine (VM) shielding for Linux VMs.
While it's easy to view this particular capability as a "nice to have," I think that from Microsoft's standpoint, it was absolutely critical to deliver this capability in the forthcoming release.
Once upon a time, Microsoft had a near-monopoly in the enterprise server market. Today, however, Linux servers account for a sizable percentage of the VMs running in datacenters around the world. From a business standpoint, Microsoft cannot afford to neglect these Linux VMs. If Hyper-V is to be a long-term success, then it has to support Linux VMs just as well as Windows VMs.
The ReFS File System
Another thing that has caught my attention about Windows Server 2019 is that Microsoft has made improvements to the ReFS file system.
ReFS was first introduced in Windows Server 2012 as a replacement for the aging NTFS file system. Microsoft's main goal behind the creation of ReFS was to create a file system that could scale to meet the needs of today's datacenters, while also providing a high degree of resiliency. In fact, the name "ReFS" actually stands for "Resilient File System." To that end, ReFS is designed to automatically sniff out data corruption and perform automatic correction while the affected volume continues to remain online.
The features that are baked into the ReFS file system for maintaining data integrity are great. There's just one small problem with the ReFS file system: Nobody is using it.
OK, I'm sure that there probably are people who use ReFS, but I have personally never seen anyone run ReFS in a production environment.
To me, the ReFS file system always felt like a work in progress. I thought that Microsoft had a great idea when it created ReFS, but I always kind of suspected that the engineers who designed it ran out of time and had to ship an unfinished component. I have absolutely no evidence to back that up -- it is just a hunch.
I am happy to say, however, that Microsoft is revisiting ReFS in Windows Server 2019, and has added data deduplication capabilities. The file system still does not support native encryption or compression, but that's OK. The addition of data deduplication capabilities will finally make ReFS a good choice for use on volumes containing Hyper-V VMs.
Microsoft has also done major work on failover clustering in Windows Server 2019.
Among the new capabilities is the ability to easily move an entire failover cluster from one domain to another. Perhaps even more importantly, Windows Server 2019 is going to make it possible to create groups of clusters, and it will finally become possible to live-migrate Hyper-V VMs between clusters.
There are two reasons why I see these new capabilities as being significant from a Hyper-V standpoint. First, I think that these capabilities lay the groundwork for eventually being able to seamlessly live-migrate VMs between our datacenters and Azure.
Second, having the ability to create cluster groups could potentially be a catalyst for a major architectural shift in the way that VM resources are arranged. I am completely speculating here (to the best of my knowledge, Microsoft hasn't talked about this at all), but we could end up with clusters that are designed to support specific VM roles.
For example, we might create clusters that are used solely for hosting Web front-end servers, or mail servers, or some other role. Similarly, we might create workload-specific clusters with all of the VMs making up a particular workload falling into a dedicated cluster.
Since we will be able to live-migrate VMs between clusters, we will presumably have the ability to architecturally structure our failover clustering infrastructure in the way that makes the most sense for our own organization's needs.
Brien Posey is a 20-time Microsoft MVP with decades of IT experience. As a freelance writer, Posey has written thousands of articles and contributed to several dozen books on a wide variety of IT topics. Prior to going freelance, Posey was a CIO for a national chain of hospitals and health care facilities. He has also served as a network administrator for some of the country's largest insurance companies and for the Department of Defense at Fort Knox. In addition to his continued work in IT, Posey has spent the last several years actively training as a commercial scientist-astronaut candidate in preparation to fly on a mission to study polar mesospheric clouds from space. You can follow his spaceflight training on his Web site.