1
00:00:00,840 --> 00:00:02,810
Here's another image of Docker workflow.

2
00:00:02,810 --> 00:00:03,100
Again,

3
00:00:03,100 --> 00:00:05,950
I provide all the attribution in the lower left corner

4
00:00:05,950 --> 00:00:08,190
here. I only take credit for my own art.

5
00:00:08,190 --> 00:00:11,910
What I want you to see here working from left to right is the developer will

6
00:00:11,910 --> 00:00:15,550
use Dockerfile. Now, it goes beyond that. You can actually create

7
00:00:15,550 --> 00:00:21,110
multi‑container setups using what's called Docker Compose, and that uses a

8
00:00:21,110 --> 00:00:25,640
YAML syntax, etc. Check the Pluralsight library, because we have plenty of

9
00:00:25,640 --> 00:00:27,740
Docker training if you want to go deeper.

10
00:00:27,740 --> 00:00:29,150
But for a single container,

11
00:00:29,150 --> 00:00:33,590
like let's say the developer wants to create an IIS web server Docker

12
00:00:33,590 --> 00:00:37,020
container. They can do that absolutely, with their Dockerfile and their

13
00:00:37,020 --> 00:00:40,340
Docker Client tools they'll create an image. And as I said,

14
00:00:40,340 --> 00:00:43,650
the developer will want to post that image to a registry.

15
00:00:43,650 --> 00:00:43,770
Now,

16
00:00:43,770 --> 00:00:47,580
Docker Hub is Docker's own public repo. You can buy

17
00:00:47,580 --> 00:00:49,670
access to create private repos.

18
00:00:49,670 --> 00:00:54,360
What we need to know for AZ‑800 is Azure Container Registry, or ACR.

19
00:00:54,360 --> 00:00:57,890
This is the Microsoft Azure‑hosted container registry.

20
00:00:57,890 --> 00:01:01,650
And as you'll see in the demo, Microsoft made ACR so that your

21
00:01:01,650 --> 00:01:05,410
developers and infrastructure people can continue to use their Docker

22
00:01:05,410 --> 00:01:07,810
tools. So you use docker push, docker pull,

23
00:01:07,810 --> 00:01:11,720
you don't have to use PowerShell or the Azure CLI, although you could use the

24
00:01:11,720 --> 00:01:16,060
Azure CLI if you want to. Microsoft wanted to empower everybody, their customers

25
00:01:16,060 --> 00:01:20,310
that is, to be able to use their existing tooling to interact with Azure

26
00:01:20,310 --> 00:01:24,990
Container Registry. Azure Container Instance is an Azure‑hosted Docker daemon.

27
00:01:24,990 --> 00:01:30,130
It presents an alternative to having the Docker Server running on a VM, let's

28
00:01:30,130 --> 00:01:34,630
say, or in your local environment. You can just spawn these containers, and even

29
00:01:34,630 --> 00:01:35,420
more to the point,

30
00:01:35,420 --> 00:01:38,580
you can spawn containers from images that are in your Azure

31
00:01:38,580 --> 00:01:42,430
Container Registry and host it all within Azure. Now Azure

32
00:01:42,430 --> 00:01:44,770
Container Instance is fine, it works great,

33
00:01:44,770 --> 00:01:47,240
but it's not necessarily enterprise scale.

34
00:01:47,240 --> 00:01:50,030
For high availability and enterprise scale,

35
00:01:50,030 --> 00:01:53,500
you've got Azure Kubernetes Service, or AKS. This is a

36
00:01:53,500 --> 00:01:56,540
Microsoft Azure‑hosted Kubernetes cluster.

37
00:01:56,540 --> 00:01:59,320
There is another option called Azure Service Fabric.

38
00:01:59,320 --> 00:02:02,080
It's very similar to Azure Kubernetes Service,

39
00:02:02,080 --> 00:02:04,580
but it's more proprietary to Microsoft.

40
00:02:04,580 --> 00:02:08,170
Microsoft appears to be de‑emphasizing Service Fabric

41
00:02:08,170 --> 00:02:10,130
and pointing customers to Kubernetes.

42
00:02:10,130 --> 00:02:10,540
And, again,

43
00:02:10,540 --> 00:02:20,000
because Kubernetes is available on‑prem and in other clouds, I would recommend that you standardize on AKS rather than Service Fabric.