(upbeat electronic music) - Coming up, we look at the
upcoming Windows Server 2019 from major advances in managing hyper-converged infrastructure, support for storage class memory, the new Storage Migration Service, as well as deduplication, coming to ReFS, new hybrid integration between Windows Server data center and Azure, with the Windows Admin Center, and much more. Now, I'm joined today, and welcome, Jeff Woolsey! (applause) - Thank you, thank you! It is great to be back. - It's good to have you back. Now we've seen some important
Windows Server milestones, from the work done in Windows Server 2016, and the semi-annual channel releases, but how do things evolve
further in Windows Server 2019? - So with Windows Server 2019, we are continuing our commitment to hybrid integration,
based on your feedback, What we just saw there. We know that many of you are focused on integrating your on-premises
data centers with the Cloud. So whether you're running
Windows Server in Azure, or on-premises in Azure Stack, or using a Windows Server
software defined solution, on-prem, we want to give
you consistency throughout. - Now, we've talked recently about the long-term servicing channel releases and semi-annual channel release, but, for people, bring people to speed. What does Windows Server 2019 fall into? - Windows Server 2019
is our next big release that uses the LTSC, or
Long-Term Servicing Channel. Remember, the LTSC is great
for your traditional apps, like SQL Server, Exchange, SAP, it's also great for
hyper-converged infrastructure. It comes with five years
of mainstream support, five years of extended support. Windows Server 2019 includes
all of the innovation that we've had in the semi-annual channel, that's all accrued here, which occurred after Windows Server 2016, and brings that into the LTSC. So this means Windows Server
2019 is a great release for both traditional applications, and new application deployments. And of course, we're focused on hyper-converged with hybrid integration, we're resimplifying
management with Admin Center, and enabling easy integration with Azure. - That sounds awesome, and lots of improvements, of course, a number of areas there, especially when it
comes to hyper-converge. Now, I know managing a
hyper-converge configuration's been a pain point for a lot of people, so what are we doing there? - Oh my goodness, we're doing a lot. Hyper-converge is a big
area of focus for us. If you look back five years or more, it used to be that you'd build
a rack starting with servers, then you'd buy some storage, whether it's iSCSI or Fibre Channel, and configure a storage
fabric, which also meant usually buying fibre channel switches. Next, you'd probably buy
some network appliances, like gateways, firewalls,
load balancers, VPNs, and add a top of rack
switch to all of those. With the hyper-converged
infrastructure solution, we create an elegant solution
using mainstream servers, ethernet switches, and software. Here, you can see servers
populated with local, fast storage, attached
via standard ethernet, coupled with the magic of software-defined compute,
storage and networking, and provide these capabilities
at a smaller footprint. And, we deliver this as
a certified, validated Windows Server Software-defined solution. - So, do you have to wait
for Windows Server 2019 to take advantage of
some of these new things? - Definitely not. Windows Server Software-defined
solutions are popular, and available right now. If you're interested today, you
can check out the WSSD page, and choose hyper-converge solutions from the dozens of options available
for Windows Server 2016, and of course, our partners
are already working with us on Windows Server 2019 solutions, and all the new features
that we're developing there. - Great, so, what are
some of those new features that we've added in 2019? - Well, let's start with storage. So, Windows Server has included granular block-based deduplication since Windows Server 2012 R2. With block-based deduplication, a file is divided into smaller blocks, and checksummed into chunks. Those identical chunks, the same checksum, are stored only once. Because block-based
deduplication goes sub-file, it's very efficient, as opposed to, say, filed-based dedupe, which generally just looks at the file name, date and size. Now previously, Windows Server dedupe has only been limited to
NTFS formatted volumes. Well, no more. Windows Server 2019 brings
dedupe to REFS formatted volumes. - [Matt] Nice. - So let me show you a demo
with Windows Admin Center. Here, we have a setup,
clustered shared volume. C:\ClusterStorage, and you can see there
are two folders here. These folders are exactly the same, except one is using dedupe-enabled,
and the other is not. So let's take a look inside one of these, and you can see it's filled
with a variety of data types. There are files, common
files of all different types. We've got about 20 virtual machines, varying in size from 10
to 20 gigabytes each, we've got SQL databases, websites, your storage probably looks
very, very similar to this. Now, let's switch on over to the hyper-converged cluster manager, where we can get a view of
everything in the cluster. I'm gonna click on volumes, and switch on over to
inventory view for details. Notice immediately, the red alert here by the volume not using dedupe, while the one using dedupe is fine. You notice they're both formatted ReFS, they're both using
three-way mirror resiliency, but notice the storage used. If you look at the storage, the one that's not using
dedupe is 92% full. So, if we then click on
view details, you can see there's further information. I'm gonna click on this inventory view, and again, the one with deduplication, you can see, guess what, it's
only 9.64 gigabytes used. If we click here for the path,
you can see the path used, as well as the total, 199 gigabytes, but again, only 5% of this
whole volume is being used. If you wanna go here into the PowerShell, and if you wanna type, "Get-DedupVolume", you can actually see, from
PowerShell, the 94% savings rate. - [Matt] Wow. - So that's dedupe, with
the resilient file system new to Windows Server 2019. - Awesome, so some big savings there, and what I'm seeing is
that Windows Server 2019 provides the best of both worlds. It's efficient, its
resiliency, performance, and deduplication, all
built in, which is great, but it's only more with storage. - Oh yes, there's a lot more. So, over the past few years, we've seen the growth of Flash storage everywhere. In addition, Flash has gotten faster, and to do that, the
industry has had to change the way Flash connects to your computer. So, first, Flash was connected via USB. But then the USB interface
simply got too slow, so Flash was connected via SATA and SAS. Guess what? Those interfaces got way too slow. So then, PCI was used, and then a whole new connection
was created, called NVMe. - And NVMe, if people are unfamiliar, stands for Non-Volatile Memory Express, which is currently the fastest Flash-based storage available. - Currently the fastest one. Now, each time we change the connection, we're moving the Flash closer
and closer to the processor. We're doing this to reduce latency, and increase performance. Well now we have memory that
plugs into a memory socket, a DIMM socket, sitting
right next to the processor, called storage class memory. Storage class memory can be
configured to look like memory, it can be treated like an
insanely fast storage device, or segmented to do both,
based on the use case. So for hyper-converged infrastructure, storage memory can be used
as an intelligent cache for storage spaces direct to drive up performance,
and reduce latency. It is amazing. - Cool stuff, and with all
these storage investments, it looks incredible. I've got a bunch of file servers, in my lab for instance, 2003, 2008. Is there a way for me to
migrate my old file servers to a new Windows Server 2019 file server, and how do I move it with confidence? ACLs, hidden files, IP
addresses, all that stuff? - This is a really tough problem. So, customers have data on file servers going back almost two decades. In many of these cases,
there are legacy apps that are looking to access a file server by IP address, or DNS. And if that changes, there could be unintended consequences,
like breaking apps. So people literally
leave these file servers, and don't touch them. What you need is a
Storage Migration Service that can take literally all of the data from a file server, migrate it to a brand new server, clone all of the identification
like DNS, name and IP, ensure all of the data is migrated, and then safely
decommission the old server. - So that sounds like
exactly what we need, also sounds like magic. - [Jeff] It does. - So, do we have a solution? - Yes, in Windows Server
2019, we're introducing the brand new Storage Migration Service. Storage Migration Service
consists of three phases. Phase one is inventory. In this case, the Storage
Migration Service orchestrator interrogates your storage,
interrogates the network, interrogates the security,
SMB share settings, everything, to understand all
the data it needs to migrate. Step two is the transfer phase. Now the admin creates
pairings from the source, to the target, and decides
what data to transfer, and performs one or more transfers. Finally, we have the cutover phase. In phase three, the admin assigns the networks to the destinations, and the new servers actually take over the identity of the old server, so it looks exactly the
same on the network, and to all of those apps out there. And so the old servers then
enter a maintenance state, where they're unavailable to users and applications for
later decommissioning, while the new server uses
the subsumed identities to carry on all duties transparently. - Sounds awesome. - So, let's take a look. And I'm giving you a world-class premiere, this is the brand new UI for
the Storage Migration Service. So here, you can see this is
all running in Admin Center. We've had a preview build available, this is the latest and greatest, and you can see those three phases are located at the top. Inventory, transfer and cutover. So the very first thing we're gonna do is we're gonna create a brand new job. And we're gonna go ahead
and give that job a name. And I'm gonna go ahead
and start with demo, and you can see I'm running all of this you're gonna see is happening
through Admin Center. So the first thing I do is I'm gonna give it my user name, my credentials for the
devices I wanna migrate, so my user name and password. And I can even include
administrative shares if I'd like. Now I'm gonna click on next, and the next thing I'm gonna do is, okay, now that I've entered my creds, I need to scan and add some devices. One of the things I
wanted to do was show you not only moving one device, we're actually gonna migrate two file servers at the same time. One of them's running 2008,
and one of them's running 2012. So we've added one, FS01, and we've added a second one, FS02. So, not only am I doing one migration, I'm actually doing two, simultaneously. So we've added the devices,
we've scanned those, we're now doing the inventory to get everything we
need to know about these. What are their shares? How many volumes do they have? What's their IP address? What's their DNS? Everything I need to know
about these old servers. And so I can see everything succeeded, I know how many files that are located, and if I click down here
below, I can see the shares. I can see all the information,
all of the shares, all of their styles, number of files, where everything is located. If I wanna see the details, I can see the path, the
shared type, I can see caching is the file encrypted,
is this a temp, volume, all of this information
is located right here. And here it is for the second server. So again, I'm not doing
one, I'm actually doing two, and if I wanna see the configuration, you can see, here's the name, I can see the network adapters. Here's the IP address. Here's the volumes on the second server, so again FS01 and FS02. So now we click on next, and now we're gonna go through the transfer phase. Here in the transfer phase,
it's time for some creds again, so here we enter the creds,
the user name and password, because now we're gonna transfer the data from the old servers to the new servers. So again, I'm doing two servers at once, and I'm adding the destination device, and this is gonna be
2019 server number two. We're gonna scan that, and what it does is now it says, "Okay, let me map the first server "to the first destination." We auto-map all the volumes, we make sure we have enough space for everything, you can see all of the shares
below, all of the information. If I decide I don't wanna
include one of these specific ones in the
transfer, I can deselect it, but I'm gonna be super safe,
and I'm gonna keep everything. And so there is everything for FS01. I click on next, and now
I'm gonna takes me through exactly the same process for FS02. Again, I'm doing this for
two servers simultaneously. So now I'm gonna, destination for this one is
gonna be new server sms2019-03, and so again, I'm scanning this, I'm auto-mapping all the
volumes, and I'm clicking next, since I showed it to you on the first one. Now we have the transfer settings, we can actually do checksums, we can configure all of
the retries if you want, but we can literally checksum every single file happening
here in the process. And we click on next, and now we go through
a validation process, because you're not gonna want to do this, unless you can be sure that
it's really gonna succeed. So we go through an extremely
stringent validation process, and if you click on this, we'll show you. We make sure that you've
got the right credentials, we make sure all the volumes are unique, we make sure we've got enough storage, we make sure the file systems match, everything we need to do to
make sure there, is right there. So we click on next, and
continue right along. Now we click on start transfer. So now, we're in the transfer phase. We are literally copying the
servers, my two old servers, one of them is running Server 2008 R2, one of them's running
2012, to my new servers. Check out this super
sexy screen right here. This is brand new. So you can see the running jobs happening, both FS01 and FS02, right now, are copying all of their
data on the network, this is all live data
happening right here, and you can see the speed
at which it's happening, about 79 megabits per second, and you can see all of the information, all of this updating in real
time in Windows Admin Center, because again, all of this is integrated for Storage Migration Service, in the Windows Admin Center console. So we can see now, guess
what, it's all done. Now of course, your boss is gonna say, "Well, how do I know that
you really got everything? "Because the guys in marketing claim "that you forgot to copy
some of their files." Well you know what? We're gonna click on the transfer log so I can show you, because you want proof? What's better proof to
anybody than to bring up an Excel spreadsheet that shows you every single file that was migrated? You want every little bit of information, 'cause your boss goes,
"Did you really copy it?" Well, here's the spreadsheet with every file, every name, every path, every type, every size, every attribute, creation, access date, modification date, owner date, ACL date,
checksum, reparse point, it is there. (laughter) So you can literally hand them
this spreadsheet that says, "Uh, yeah, we did. "We're good." Now, we have the final phase, cutover. So the cutover phase is now
where we're basically saying, "Okay, these new servers, "they're gonna take on the IP address "of the old servers." These are gonna look exactly like the old servers on the network. So we're gonna type in, we're gonna give the ethernet address, and it's gonna map the new IP addresses so they look exactly the same. It's gonna change their
names, so the new servers are gonna be named FS01 and FS02. We are doing all this
work under the covers for you to do this cutover process. So I configured the
cutover for number one, here's the configure, the
cutover for number two, and again, all this work is click, click, click, and validating everything along the way. Now we go through the cutover settings, and now again, we go through
another validation process, 'cause we wanna make sure that you are 100% confident
that everything works. So we go and validate both of
the source and destinations, and again, we're gonna get
a link on over here to see, guess what, yup, the
first one, FS01 is passed, and lo and behold, the second
one, FS02 is passed as well, and if I wanna click, all the information I
need right there, yup. Everything passed, I have
all the names, credentials, passwords, storage,
everything is ready to go. Guess what, we're gonna click on next, and now we're gonna go through
the final cutover process. And in this last cutover
process, by the way, if you want to sync
this storage with Azure, and Azure File Sync, we also give you a
little bit of information at some point, if you
decide you just wanna move some of this to Azure,
you can, and tier it. But in the meantime, we're
gonna start this cutover, this a little time-compressed right here, but right now, this is all cutting over, all on-premises, and I'm
going from that ancient server that had spinning discs,
to brand new servers filled with Flash, and
I'm probably getting a 10 to 20 to 50x performance improvement, going from spinning rust to Flash. And so, that is the
entire cutover process. And so now if I wanna
go back on over here, and I wanna take a look
at the cutover process, you can see everything succeeded, and that is the whole process right there. That is the new Storage
Migration Service in 2019. How'd we do, guys? (applause) And, spoiler alert, we go all
the way back to Server 2003. So for those people
that raised their hand, and said they had Server
2003 file servers, we actually go all the
way back, and can migrate from 2003 on to a modern file server. We actually did all the work, congratulations to Ned Pyle
and the team, to do that, so we can help you move those really ancient legacy servers forward. - Nice work, Jeff, that was awesome. And you even managed to ignore
your meeting notification... - I don't have time
for meetings right now. - (laughs) That was great. Now, I love the fact that there's no risk to the old file server, as you said, if people are running legacy systems, this is a great way to transition. The whole operation is a non-destructive migration.
- Non-destructive. - Okay, let's switch gears to management. - [Jeff] Yes. - How easy is it to manage
a hyper-converge cluster? - Three words for ya. Windows Admin Center. So as you just saw, Admin
Center is the future of Windows Server management, and we are investing heavily there. In fact, notice the last demo. Everything was done in storage migration, and you're gonna see it all week long. So let's see Admin Center in
action for hyper-converge. So let's start by taking a look at this hyper-converged cluster
demo cluster right here. I can click on the dashboard, and you can see, immediately, keep in mind, this isn't
looking at a single server. It's looking at an entire cluster. So you can see, I have a
small three node ACI cluster that is still very powerful. I see the alerts, and the overall cluster
storage performance. Here you can see it's around 670,000 IOPS, with about 50 microseconds of latency, and a throughput of 2.6
gigabytes per second. If I want, I can even go back over here and mouse over specific
points if I wanna see real time data that's
happening at specific times. Also notice up here in
the right hand corner, that we keep and maintain historical data. So I can click on day to
see the last 24 hours, and I can look for any
interesting peaks and valleys. You'll quickly find out when
people show up into the office. Or, if you want, you can
specify the past week, month, up to a year. Next, you can look at resource utilization for the entire ACI
cluster, and see CPU usage, memory, and storage,
all in one quick glance. Notice how quickly you can just... It all just makes itself
readily available. In terms of servers, I can see all my three servers are
healthy in this cluster. I have 18 drives, all 18 of them, there are six in each
server, six per node, they're all healthy. You can see I have a total
of 17 virtual machines, 14 are running, two are paused, and I have 12 volumes. One is critical, and one has a warning, I'm gonna need to take a look at that right after the demo here. - Nice. Lots of cool stuff. Really great hyper-converge management. Now, what about DR? What are we doing to enable
hybrid scenarios there? - So again, when it comes to hybrid, Windows Admin Center is key
to our hybrid integration. We've had virtual machine
replication for many, many releases, so you can replicate VMs between your sites, or up to Azure. However, many of you don't
have a secondary site, or some of you that do have
one would prefer to retire it, because it costs a lot, and so you'd prefer to use
Azure as your DR target. Well, with Admin Center, we've made business continuity
easier than ever before. So, here I am in Windows Admin Center in the virtual machines
tool, with a list of VMs. And I'm gonna replicate this
first one, Hono17093, to Azure. And so the first thing
I do is select the VM, and I go to the more settings
to click set up VM protection, because I haven't set this up yet. First thing I do is, I'm
gonna associate this VM with the Azure site recovery, so I'm prompted to sign in to Azure. I give it my creds and my password, and now I give it my
subscription information, my subscription, my location, as well as create a new storage account, and my vault, and that's it. This is a one time
process, I've set up ASR. That's it. Now, you notice up in
the right hand corner, there's a notification letting me know registration is happening, and when it's successfully
set up, just like that. So now, if I go back
to the VM select more, and click protect VM, that's it, I'm done. All I have to do now is
select my storage account, and just like that, you can see up here in the right hand corner,
we are enabling protection, and the protection status
has gone into replicating. So we're now replicating
this VM up into Azure. - Awesome, cool stuff. Now, what does it look like, actually in Azure, when
that VM gets there? - So if we wanna switch
on over to Azure here, you can see, if we go to
the recovery services vault, and go to that storage
one I just created here, you can see, from the subscription, that in fact, here is this VM, and in fact, it is replicating
and healthy right there. - Awesome stuff, nice work, Jeff. So the combination of Windows Admin Center and Azure provides a really
powerful hybrid solution. Everyone'll agree. Are there any limitations
around the guest? Does it need to be Windows,
or can it be Linux? - No, it can be Windows, it can
be Linux, it doesn't matter, it is completely transparent to the guest. - Okay, great. Well, thank you very much, Jeff. Great overview of the latest
updates to Windows Server 2019. Thanks for joining us so much today, Jeff, and where can folks watching learn more? - So, join the Windows
Server Insiders Program. You get access to new
builds every couple weeks, and that includes Admin Center
along with Windows Server. Also check out and download Admin Center. If you haven't done that yet, you need to, because it works right now with the servers you have right now. It'll manage all the way back to 2012, with some limited support for 2008 R2. - Awesome, thanks Jeff. And of course, keep checking
back on Microsoft Mechanics for the latest tech updates. Thanks for watching, everyone. (applause) (upbeat electronic music)
