Kubernetes has Crossed the Chasm: The Platform is Too Big to Ignore.

If you want a glimpse at what Kubernetes is, why it's used, and what the benefits of Kubernetes are, this page will give you just that.

We recommend that you also check out the ebook to find out how companies of different sizes are using Kubernetes to plan for the future.

What is Kubernetes, and why is it used?

Kubernetes, sometimes called K8s, is an open-source system used to automatically deploy, scale, and manage containerized applications across a cluster of hosts, on-premises, or in the public cloud.

In short, containerized applications run in isolated runtime environments known as containers. These hold an application and all of its dependencies, like system libraries, binaries, and configuration files.

Thanks to this all-in-one packaging, the application becomes portable and will behave consistently across hosts, which enables developers to write once and run the application almost anywhere.

Kubernetes streamlines application management by

1. Automating operations that are associated with container management

2. Providing built-in commands for application deployment, the rollout of updates, scaling up and down to accommodate changing requirements, monitoring, and more.

The result? Kubernetes makes it easier to manage applications.

In the sections below, we dive more into the basis of Kubernetes architecture and deployment evolution while also looking at typical use cases.

Towards the end, we've linked several Kubernetes resources and how to start your Kubernetes journey. If you immediately want to learn how other companies use Kubernetes, download the ebook via the form above. 

Why do we need Kubernetes?

Before we continue to explore Kubernetes, let's answer the question: why do we need Kubernetes?

Traditional deployments

In the past, businesses used physical servers to operate. The issue was that while numerous actions were being carried out on a single server, one application could suck up most of the resources and degrade the performance of other operations.

Adding extra servers was one option, but as you can imagine, this quickly got quite expensive.

Virtualization deployments

Then, everything began to move towards virtualization. A single physical server's CPU could support many Virtual Machines (VMs), which allowed multiple programs to run concurrently without experiencing performance difficulties.

Moreover, VMs allowed for the isolation of applications, which added an additional layer of flexibility and security. The entire system was not impacted when one program needed to be added, updated, or fixed. Yet, a significant problem with VMs was their high memory utilization.

Container deployments

Meet containers. Containers and VMs are much alike. They are independent of the central infrastructure and have their own operating systems, file systems, CPU, memory, and process spaces.

The main difference is that containers have a significantly smaller memory footprint thanks to their relaxed isolation properties.

Today containers have, in some cases, replaced virtual machines. Containers run a variety of intricate application clusters, which are frequently complex to manage effectively.

And it's here that Kubernetes comes in. The image above illustrates the deployment evolution from traditional to container.

At its core, Kubernetes and other container orchestration tools support developers through complex application management. Developers take care of the app management to prevent downtime.

For example, imagine a container fails; now, another container must replace it. But rather than doing this manually, Kubernetes manages this switchover seamlessly and effectively by restarting, replacing, and removing failed or stalled containers that don't respond to health checks.

What are the benefits of Kubernetes?

There are many reasons to consider Kubernetes, but here are three highlights:

Kubernetes automates operations

You can automate daily processes using Kubernetes, as it comes with built-in commands that take care of many manual and labor-intensive aspects of application management.

That way, you can ensure that applications always operate the way that you planned.

Kubernetes keeps the underlying environment in check

After the installation of Kubernetes, it will manage the computing, networking, and storage system on behalf of your workloads.

That means developers can now focus on applications rather than the underlying environment.

Kubernetes provides better resiliency

Kubernetes repeatedly monitors the health of your services. For example, if a container fails or stalls, Kubernetes restarts it and only makes the service available to users after verifying that it's running.

The basis of Kubernetes architecture

Kubernetes architecture falls under several names, like Kubernetes application deployment or Kubernetes client-server architecture.

No matter the name you use, the basis of the architecture is that it's used to compose, scale, deploy, and manage containerized applications across host clusters.

The essential elements of a Kubernetes environment are:

• A control plane,
• A distributed key-value storage system (to ensure clusters remain steady), and
• Cluster nodes (Kubelet, or worker nodes).

Now, a Kubernetes cluster is a type of Kubernetes deployment architecture. The control plane and the nodes, or compute machines, comprise the basic Kubernetes architecture.

Each node has its own Linux environment and can be a physical or virtual machine. Also, each node runs pods, which are made up of containers.

The Kubernetes control plane and the cluster's nodes are included in the Kubernetes architecture. And inside the control plane machine, you will find The Kubernetes API server, Kubernetes scheduler, Kubernetes controller manager, and etcd.

The Kubernetes node elements comprise a container runtime engine (also known as a docker), a Kubelet service, and a Kube proxy service.

The architecture of Kubernetes clusters

We briefly touched upon Kubernetes clusters, but let's zoom in more.

Cluster nodes are computers that run containers and are controlled by the control plane. Each node has a running agent for interacting with the main Kubernetes controller, the Kubelet, which serves as the control plane.

A container runtime engine, such as Docker or rkt, is also run on each node. The node also utilizes additional components for service discovery, monitoring, logging, and optional extras.

Kubernetes nodes

There must be at least one computing node in a Kubernetes cluster, though there may be more depending on the demand for capacity.

In addition, more nodes are required to scale and increase cluster capacity since pods are coordinated and scheduled to operate on nodes.

Nodes do the work for a Kubernetes cluster. They link apps and networking, compute, and storage resources. Nodes could be bare metal servers in data centers or cloud-native virtual machines (VMs).

Container runtime engine

A container runtime engine is used by each compute node to operate and control container life cycles. Open Container Initiative-compliant runtimes, including Docker, CRI-O, and rkt are supported by Kubernetes.

Kubernetes pods

Up to this point, we have focused on internal and infrastructure-focused concepts. Pods are different. Pods are the main external component that developers work with.

For example, in the Kubernetes object model, a pod represents a single instance of an application and is the most basic object.

Pods, however, are the core and foundation of Kubernetes. Each pod is made up of one or more tightly coupled containers arranged in a logically sequential order, as well as rules that govern how the containers operate.

Kubernetes Resources: Ebook

If you're looking to get started with Kubernetes and learn how other companies are deploying the platform, we recommend downloading the ebook, The State of Kubernetes. In short, the ebook

• Examines the continuing evolution of how enterprises are using Kubernetes.
• Discusses if the challenges this advanced technology represents are worth it.
• Breaks down the benefits and speed bumps that Kubernetes brings based on responses from 357 qualified software development and IT professionals from various roles, regions, industries, and job levels.

Kubernetes Resources: Glossary

You may also want to check the complete VMware glossary on Kubernetes, diving into Kubernetes services, the difference between a service and a deployment, the components of a Kubernetes service, and how it works.

Getting started with Kubernetes

And if you just can't wait to test the waters and test Kubernetes for your workloads, check out the guide from Tech Target: How to deploy Kubernetes on VMware with vSphere Tanzu.

Of course, our team is always here to support you as well, so feel free to reach out to our technical consultants.