1
00:00:00,940 --> 00:00:04,630
Now a stretch cluster is kind of a special case scenario.

2
00:00:04,630 --> 00:00:09,480
The idea here is that you have multiple sites and you need to extend a

3
00:00:09,480 --> 00:00:12,940
single cluster or stretch a cluster across those sites.

4
00:00:12,940 --> 00:00:15,240
Now depending upon how you're connecting to Azure,

5
00:00:15,240 --> 00:00:19,310
if you have an ExpressRoute or a site‑to‑site virtual private network,

6
00:00:19,310 --> 00:00:21,190
you could stretch in a hybrid cloud.

7
00:00:21,190 --> 00:00:25,360
But I've seen in industry more often it's physical site to physical

8
00:00:25,360 --> 00:00:29,780
site. But as long as you have the bandwidth and capability to do this,

9
00:00:29,780 --> 00:00:33,670
notice that what you've got physically is a separate shared storage

10
00:00:33,670 --> 00:00:38,610
array in each site and you've configured storage replica to replicate

11
00:00:38,610 --> 00:00:40,840
that storage to keep it in sync.

12
00:00:40,840 --> 00:00:43,750
And then when we define the stretch cluster,

13
00:00:43,750 --> 00:00:45,520
you'll see that we've got in this picture,

14
00:00:45,520 --> 00:00:47,890
Node1, 2, 3, 4, 5, 6.

15
00:00:47,890 --> 00:00:51,040
They're all voting nodes in the same cluster.

16
00:00:51,040 --> 00:00:53,240
And because of that storage replication,

17
00:00:53,240 --> 00:00:59,540
we could, for example, failover that VM on Node2 to Node4.

18
00:00:59,540 --> 00:01:03,230
So it just provides another layer of abstraction for high

19
00:01:03,230 --> 00:01:07,340
availability with your cluster service.

20
00:01:07,340 --> 00:01:10,930
What is the Windows Server failover cluster deployment process?

21
00:01:10,930 --> 00:01:13,010
Well, you always want a plan.

22
00:01:13,010 --> 00:01:15,580
You want to think about how many nodes you have now,

23
00:01:15,580 --> 00:01:17,400
how many nodes you might have in the future,

24
00:01:17,400 --> 00:01:18,830
your need for a witness.

25
00:01:18,830 --> 00:01:21,690
What kind of roles are you planning to cluster?

26
00:01:21,690 --> 00:01:24,440
What have you decided on at the storage layer?

27
00:01:24,440 --> 00:01:27,800
Are you going to do a SAN, a Fibre Channel, iSCSI?

28
00:01:27,800 --> 00:01:32,130
Or are you going to do a Storage Spaces Direct shared storage,

29
00:01:32,130 --> 00:01:33,800
you know, those sorts of questions.

30
00:01:33,800 --> 00:01:37,420
Then we install the appropriate server features on your

31
00:01:37,420 --> 00:01:39,920
nodes that will participate on the cluster.

32
00:01:39,920 --> 00:01:44,350
It's strongly recommended that you run the cluster validation wizard to

33
00:01:44,350 --> 00:01:47,140
make sure that you don't have any blocking issues.

34
00:01:47,140 --> 00:01:48,960
Then you create the cluster.

35
00:01:48,960 --> 00:01:52,520
The cluster also has its own client access point,

36
00:01:52,520 --> 00:01:57,940
so there'll be a virtual IP address and DNS name for the cluster itself.

37
00:01:57,940 --> 00:02:00,870
You'll then want to configure your storage, however you're doing that.

38
00:02:00,870 --> 00:02:04,580
This is your shared storage that's connected to all cluster nodes.

39
00:02:04,580 --> 00:02:11,000
And then, lastly, you deploy the cluster roles and then monitor ongoing. So that's the high‑level view.