WEBVTT 00:00.670 --> 00:05.770 Hello and welcome to this lecture. In this lecture we talk about the different choices available 00:05.770 --> 00:11.440 for the infrastructure hosting a Kubernetes cluster. In the previous lecture we discussed about the various 00:11.440 --> 00:13.870 options we have in deploying a kubernetes cluster. 00:13.900 --> 00:17.140 Let's look at it in a bit more detail here. 00:17.170 --> 00:22.570 Kubernetes can be deployed on various systems in different ways. Starting with your laptops, to physical 00:22.570 --> 00:26.660 or virtual servers within an organization as well as those in the cloud. 00:26.860 --> 00:31.750 Depending on your requirements your cloud ecosystem and the kind of applications you wish to deploy 00:32.140 --> 00:36.070 you may choose one of these solutions. On a laptop or a local machine, 00:36.070 --> 00:38.130 There are a number of ways to get started. 00:38.140 --> 00:43.540 First of all on a supported Linux machine you can get started with installing the binaries manually 00:43.570 --> 00:45.550 and setting up a local cluster. 00:45.550 --> 00:48.890 However that is too tedious especially if you are just getting started. 00:49.300 --> 00:55.140 So relying on a solution that automates all that will help in setting up a cluster in a matter of minutes. 00:55.270 --> 00:58.810 We will look at some of those solutions in a bit. On Windows 00:58.810 --> 01:04.240 on the other hand, you cannot setup kubernetes natively as there are no windows binaries. 01:04.300 --> 01:10.780 You must rely on a virtualization software like Hyper-V or Vmware workstation or VirtualBox to create 01:10.810 --> 01:13.950 Linux VMs on which you can run kubernetes. 01:13.960 --> 01:19.570 There are also solutions available to run kubernetes components as docker containers on windows VMs. 01:19.900 --> 01:26.920 But remember, even then the docker images are Linux based and under the hoods they run on a small Linux OS 01:26.950 --> 01:31.500 created by Hyper-V for running Linux docker containers. 01:31.540 --> 01:36.040 So what are some of the solutions available to easily get started with kubernetes on a local machine? 01:36.450 --> 01:42.310 Minikube, deploys a single node cluster easily. It relies on one of the virtualization software like Oracle 01:42.310 --> 01:47.350 Virtualbox to create virtual machines that run the kubernetes cluster components. 01:47.380 --> 01:52.330 We have seen this in the beginners course. The kubeadm tool can be used to deploy a single node or a 01:52.330 --> 01:53.480 multi node cluster 01:53.560 --> 01:59.950 real quick before this you must provision the required hosts with supported configuration yourself. 02:00.010 --> 02:05.530 So the different between the first two and kubeadm is that the first two provisions a VM with supported 02:05.530 --> 02:12.910 configuration by itself, whereas kubeadm expects the VMs provisioned already. At the same time it allows 02:12.910 --> 02:18.280 for deploying multi-node clusters, whereas the former doesn’t. Again deploying a kubernetes cluster 02:18.280 --> 02:24.670 locally on a laptop is usually for learning, testing and development purposes. For production purposes, 02:24.670 --> 02:30.430 there are many ways to get started with a kubernetes cluster. Both in a private or a public cloud environment. 02:30.460 --> 02:37.330 I would categorize them as Turnkey solutions or Hosted or managed solutions. Turnkey solutions are where 02:37.330 --> 02:43.210 you provision the required VMs and use some kind of tools or scripts to configure kubernetes cluster 02:43.330 --> 02:44.040 on them. 02:44.140 --> 02:49.630 At the end of the day you are responsible for maintaining those VMs, and patching them and creating,upgrading 02:49.630 --> 02:56.200 them etc. But cluster management and maintenance are mostly made easy using these tools and scripts. 02:56.380 --> 03:03.970 For example deploying a kubernetes cluster on AWS using the KOPS tool. Hosted solutions are more like Kubernetes 03:04.030 --> 03:09.760 as a service solution, where the cluster along with the required VMs are deployed by the provider and 03:09.760 --> 03:13.220 kubernetes is configured by them by the provider. 03:13.240 --> 03:15.210 The VMs are maintained by the provider. 03:15.220 --> 03:21.040 For example, Google Container Engine lets you deploy a kubernetes cluster in a matter of minutes, without 03:21.040 --> 03:27.620 you having to perform any configuration by yourself. Let us look at some of the Turnkey solutions. 03:27.620 --> 03:31.010 OpenShift is a popular on-prem kubernetes platform 03:31.050 --> 03:36.680 by RedHat. For those of you who may not be familiar, OpenShift is an open source container application 03:36.680 --> 03:39.700 platform and is built on top of kubernetes. 03:39.770 --> 03:45.920 It provides a set of additional tools and a nice GUI to create and manage kubernetes constructs and 03:45.950 --> 03:49.640 easily integrate with CI/CD pipelines etc. 03:49.910 --> 03:53.000 We have an OpenShift for beginners course in our catalog. 03:53.070 --> 03:55.190 Check it out if you are interested! 03:55.190 --> 04:00.740 Cloud Foundry Container Runtime is an open-source project from Cloud Foundry that helps in deploying 04:00.740 --> 04:06.540 and managing highly available kubernetes clusters using their open-source tool called BOSH. 04:06.560 --> 04:11.930 f you wish to leverage your existing Vmware environment for kubernetes, then the Vmware Cloud 04:12.130 --> 04:15.910 PKS solution is one that should be evaluated. 04:16.070 --> 04:21.440 Vagrant provides a set of useful scripts to deploy a Kubernetes cluster on different cloud service 04:21.440 --> 04:22.880 providers. 04:22.880 --> 04:29.000 All of these solutions makes it easy to deploy and manage a cobra that is cluster privately within your 04:29.000 --> 04:30.090 organization. 04:30.110 --> 04:34.720 You must have a few Virtual Machines with supported configurations in place. 04:34.760 --> 04:38.470 These are few of the many kubernetes certified solutions. 04:38.570 --> 04:44.420 Well there are many more, so check them out in the kubernetes documentation page. Let us look at some 04:44.420 --> 04:50.180 of the hosted solutions. Google Container Engine is a very popular kubernetes as a service offering 04:50.270 --> 04:55.640 on Google Cloud Platform. Openshift online is an offering from RedHat where you can gain access to a fully 04:55.640 --> 04:57.910 functional kubernetes cluster online. 04:58.460 --> 05:04.770 Azure has Azure Kubernetes Service. And finally Amazon Elastic Container Service for Kubernetes 05:04.820 --> 05:09.910 is Amazon’s hosted kubernetes offering. Again, these are just some of the solutions 05:09.950 --> 05:12.710 there are many more. So what is our choice? 05:12.710 --> 05:17.870 In our case since this is for learning purposes and considering the fact that some of you may not have 05:17.870 --> 05:24.620 access to a public cloud account and since most of you mentioned in the poll we sent out that you prefer 05:24.650 --> 05:27.000 a local setup with Virtualbox, 05:27.080 --> 05:33.470 we chose to deploy a local kubernetes cluster from scratch on our local system by creating several 05:33.470 --> 05:35.520 virtual machines on VirtualBox. 05:36.350 --> 05:43.280 So our design now has 3 nodes, 1 master, 2 worker, to be deployed on a laptop with Virtual Machines 05:43.280 --> 05:45.630 provisioned on VirtualBox. 05:45.920 --> 05:47.810 Well that's it for this lecture. 05:47.810 --> 05:52.190 Check out the reference page to read more about many more such solutions.