Breaking IT Down: What Is Kubernetes?

Kubernetes is an open source tool used for automating and managing container operations. Ever since its launch as an open source platform in 2014, Kubernetes has grown to become the preferred choice for container orchestration. It has the support of tech majors and is the most popular open source project after Linux among the open source community of developers.

Container technology enables packaging of an application together with all its dependencies, such as OS, SSL libraries, and configuration files, into a single container. This technology is not new; it has been a part of Linux operations for years.

Developers have been quick to adopt container technology because it enables migration of software with all its dependencies as a single unit to a test server, or a cloud server, or a physical machine at a corporate data center, thereby doing away with compatibility issues. Container software also makes it possible to break a program into components known as micro-services.

Fragmentation into micro-services enables multiple teams to work on the components separately, so long as interoperability between different containers is maintained. This speeds up the development process. Also, container technology enables you to control containers down to the finest detail.

Origin story

Kubernetes was initially conceptualized and developed by engineers Craig McLuckie and Joe Beda at Google years ago, at a time when Google was already using Linux container technology to deploy applications. In the early days, Google used their in-house platform, Borg, to manage container orchestration. Borg managed container scheduling; it had the capacity to recognize and determine when to launch a container and when to shut it down.

McLuckie and Beda wanted to develop a more advanced orchestration layer for managing applications in a production environment. Kubernetes was born out of this desire to create a tool that would enable developers to create distributed applications and companies to manage workloads in a production environment.

Today, Kubernetes is the most widely used container orchestration platform. It offers the ease of PaaS, as well as the flexibility of IaaS, while overcoming some of the shortcomings of both.

Google open-sourced Kubernetes in June 2014. In 2015, the Cloud Native Computing Foundation (CNCF) was formed as an organization for cloud native computing and a custodian of Kubernetes. At the KubeCon + CloudNativeCon 2019 Europe conference, a survey conducted by Platform9 found that 47 percent of the 500 attendees already had some workloads running on Kubernetes in a production environment.

What you can do with Kubernetes

Kubernetes enables orchestration of production workloads. In a production environment, applications are packaged in multiple containers that need to be deployed on several hosts. Security is implemented across multiple layers. The more containers you have, the more complex managing containers becomes.

Using Kubernetes you can automate many of the manual processes needed to deploy and scale containers. Multiple hosts spanning public, private, and hybrid clouds can be grouped into clusters. By enabling efficient management of clusters of containers, Kubernetes frees you from having to manage individual containers separately. Kubernetes provides a platform for running and scaling large clusters of cloud-native containerized applications.

Once you've implemented a container-centric production environment using Kubernetes, you can orchestrate computing, storage, and networking operations at the container level across several hosts. Since Kubernetes is open source, it enables portability across different cloud infrastructure service providers.

How it is used

Use of containers ranges from basic to advanced. At the primary level, containers are faster and resource-saving versions of VMs. The complexity arises when you need to scale and run multiple applications in a production environment. This requires concurrent functioning of many co-located containers. Complexity increases as containers proliferate.

Kubernetes simplifies this, however, by categorizing containers into pods that run on nodes. This makes it much easier to schedule workloads, and manage storage and networking requirements. With Kubernetes, you can balance loads across these pods in order to make sure the number of containers in operation at any given time is just what is needed for delivering workloads, no less no more.

Common Kubernetes terms include:

Cluster: This refers to a group of machines that run Kubernetes. Each cluster has a master and nodes.

Master: This system controls the nodes and all other machines in the cluster, and assigns tasks.

Node: Nodes are physical machines or VMs controlled by the master and that execute assigned tasks.

Pod: A pod is a group of one or more containers deployed on a node. Pods are run by nodes.

Kubernetes is implemented on top of an OS. A Kubernetes Master sends instructions from an administrator to the nodes. The orchestration system can be configured to automatically identify the node best suited for the process, allot resources, and assign pods for the task.

In order to orchestrate all components of container infrastructure, Kubernetes needs to be implemented with other open source projects, such as Atomic Registry, Ansible playbooks, Prometheus, Elasticsearch, Kibana, and OpenvSwitch, to name some. For security, open source projects that are deployed include RBAC, SELinux, and LDAP.

Containers make more efficient use of resources than VMs. While each VM needs a separate OS, multiple containers can share a host OS.

Kubernetes skills in demand

Kubernetes growth has created a need for engineers with Kubernetes expertise. Currently, demand for Kubernetes skills exceeds supply. As of June 13, Indeed had 10,404 open Kubernetes-related job postings. According to Dice.com, Kubernetes was the skill most requested by IT companies in 2018.

Employers are primarily looking for professionals who have the skills to work as Kubernetes administrators and Kubernetes application developers. An application developer needs to be able to design, develop, monitor, and repair scalable cloud native applications. He should know at least one programming language, such as Java, Python, Go, or Node.js.

To know what Kubernetes skills are currently in demand, read through recent job descriptions that come up in a search for "Kubernetes skills" on leading job sites, such as Dice, LinkedIn, Glassdoor, and Indeed. DevOps developers and engineers are the most in-demand Kubernetes-related roles.

Training

IT professionals looking to develop Kubernetes knowledge and skills have a number of training options available. Certifications help demonstrate expertise for specific Kubernetes-related roles. The Linux Foundation and CNCF jointly offer the following certifications:

Certified Kubernetes Administrator (CKA)

This credential validates hands-on expertise required to function as a Kubernetes Administrator. To earn the CKA, you need to pass a 3-hour online, proctored exam consisting of performance-based problems that need to be solved from a command line.

CNCF does not specify any prerequisites for the exam. To train for the exam, you may take the self-paced online Kubernetes Fundamentals course offered by CNCF or sign on for a training program with a Kubernetes Training Partner. For additional details click here.

The certification is valid for 3 years. Additional information can be found here and here.

Certified Kubernetes Application Developer (CKAD)

The CKAD credential demonstrates the holder's ability to "design, build, configure, and expose cloud native applications for Kubernetes." To earn the CKAD, you need to pass a roughly 2-hour online, proctored exam comprising problems that must be solved from the command line.

Details are available online.

CNCF offers a self-paced online Kubernetes for Developers course, which covers the objectives of the CKAD exam.

You can also get started with Kubernetes by working with Kubernetes communities on GitHub and GitLab. As June 14, there were 27 open Kubernetes projects on GitHub. Kubernetes, with 6,500 contributors, was one of the top 10 open source projects on GitHub according to GitHub's 2018 Octoverse report.

Joining the growing open source Kubernetes community on GitHub is a great way to learn from experienced developers, gain hands-on experience, build your network, and make contributions that can add value to your resume.

CNCF also offers a free course in collaboration with edX. This course is designed for beginners and covers fundamental concepts, Kubernetes architecture, and the approach to managing deployment of containers at scale.

The KubeCon + CloudNativeCon conference held in Seattle in December generated a lot of valuable content that is now available online. With demand for Kubernetes administrators and developers rising steadily, this is an opportune time to gain Kubernetes expertise and advance your career prospects.