So within the virtual data center,
she wrote the word virtual up here is a reminder
I have, uh, storage.
of my virtual hosts.
and all of the hosts that air in this virtual data center
The service is depending on how you got everything configure.
And if we think about
what else is happening in this virtual data center,
we have to consider the fact
take a shortcut and say that each BM is composed of some applications and a guest OS.
Now, I haven't drawn any connections between any of these machines and the PM's.
But as we'll see when we go through more of a lecture and some of the labs,
we can configure these virtual machines to reside on any one of these hosts, for instance, of combination of pulling host together to make a cluster,
being able to move virtual machines from one host to another, depending on different conditions and different rules that we define.
And so that gives a lot of flexibility for managing your applications in the virtual data center.
So one of the things we have to think about is differences between for virtual machines and physical machines.
Physical machines are
what you're traditionally used to where you go into your data center. You've got a big rack with Oliver servers loaded
with a blade servers or perhaps ah, larger, dedicated raid units underneath a big server that's taking up several units in the rack.
And the physical server has its own physical hardware cards, its own attached storage and memory. Of course,
if we want to do work with a physical server
for maintenance or for certain other kinds of reconfigurations,
uh, we often have to incur downtime.
So that means that if I want to add memory to a
a server that I've built, I have to take the system down. I have to schedule that and get a coordinated with everyone who needs the applications running on that system.
When that servers down, the applications that were that were there are no longer available.
If I want to add processors to that physical serve, I have to do the same thing. I need downtime,
so that's a big limitation
in changing the hardware or the configuration of the server.
operating systems. Windows in particular require a reboot when you make certain changes, too,
the operating system or if you install patches.
That doesn't change much in the virtual world for the guest OS, in any case,
but we do have some more flexible options
now if we contrast that with a virtual machine,
a virtual machine, I can change in some cases memory processors on the fly without having to do any reboots.
You can add memory of decreased memory and more storage. Add more network cards
without having to do any rebooting. And that's a beautiful thing.
Some operations with virtual machines do require a reboot because there are certain limitations in the way that
the virtual hardware can talk between the
virtual machine and the physical host.
So we'll discuss some of those differences once we get a little further along,
if I can make changes to these hosts to some degree and changes to the to the virtual machines that are running on that host,
that means that I've got a lot more options for dealing with things like disaster recovery
and being able to move things around as needed to satisfy changing requirements in the environment.
So it stands to reason, by the way, I've drawn this diagram in this diagram
that the resources are being shared, right.
I've got network cards. I've got memory, I've got storage.
All the virtual machines running on this particular host are sharing those resource is,
and we'll see that that offers another layer of flexibility because you can
designate certain virtual machines to have more resources than others.
For instance, if this host
eight gigabytes of RAM,
I could divide this up
by giving each of these
the EMS two gigabytes each.
That's that's a very basic way to do this. Most likely you wouldn't do. You wouldn't allocate all eight gigabytes of your available ram to the BM because that wouldn't be much left for the host operate with. But you could do something pretty close to this
or I could do something like this. I could say one gigabyte for Windows seven, and because I've got a server here, maybe I want to give that when three gigabytes.
I'm still using eight, but I changed the balance,
and that gives you the flexibility I mentioned earlier for
putting the resource is where you need them to give the performance that are required for different types of applications and different types of operating environments.
We'll see a little bit later how some of the other features allow more flexibility, even even yet.
For instance, the concept that I could actually allocate more memory than I have available is a possibility within a virtual machine environment. You can't do that in a physical environment.
If I was running this host with eight gigabytes of RAM
and I wanted to use applications that required 10 gigabytes of ram, let's say altogether I would have problems that that second application or the third application then pushed me over the aching goodbye limit would not be able to start up
because there wouldn't be any memory available
in the virtual machine environment. However
I could, I could perhaps
allocate two gigabytes here
So this adds up to 24
7 10 That's tanking Good bikes total
and because of the way that the
the resource sharing operates
Yes X, I envy spear together work so that
the total amount of required. Memory doesn't have to be available at all times. If it was required at any one given moment, you would have some issues and you suffer some performance problems.
But in general, being able to over allocate your memory compared to what the host thinks it has is a possibility.
One of the one of the more advanced features of using
a virtual environment like like we're seeing here.
the applications and operating systems
that are running if we refer back to this side
virtual machines has the real memory of the allocation, as we saw here.
And I can control that dynamically by changing some of the configuration settings within the host.
Ah, and within some of the
other advanced service is that we'll see in some later labs.
We have different file system types that will be covering.
if you're considering an NFS file system,
that could be just a regular Windows partition that shared from Windows Server.
okay also be shared from a Lennox machine. So you have different operating system. Sorry, different file system choices in that case,
or if we attached to a nice cosy London.
Then we can create a
or an *** or fibre channel. We can create a BM wear file system,
and the one will be using
is the M. F. S. Five.
We'll cover that some of the differences in the operating systems
For instance, if I had a NFS data store,
I'm going to just draw connecting line
so we can move it over here.
So here's my NFS data store.
And if I think about the fact that this is consolidated storage and there might be,
uh, one physical disk that's comprising this day in store, let's let's say it's 10 gigabytes
of storage. Actually, it's not very much. We'll call it 100 gigabytes, cause if I have four b EMS, I probably would need
something more like 100 gigabytes.
But I can have some portion of the storage
that's allocated to Window seven,
another portion that's allocated to
and then maybe another portion allocated to Lennox.
So I'm sharing storage.
Connecting to it through the network
are the virtual network connector through the physical Network connector on the host
within a V M F s style system.
We have, For instance, if this if this is a V M F s,
you have the root of the file system,
which is much like the root file of a,
NFS file system on a Windows box, like the root of your C drive, for instance, or the root of a Lennox partition.
or volumes designation that shows
all the details for the configuration of that particular bit of storage.
the V M. F s structure,
then map to some other kind of physical storage.
In this case, maybe it's NFS.
It could be a local disc,
So what this means is I've got tremendous flexibility for where I allocate the storage, different storage technology that's used
a lot of flexibility for being able to grow and shrink
those storage volumes as the requirements change in your environment.