DEV Community: Stephen Rodriguez

{ Zero to Helm }: Part 2 - Architecture

Stephen Rodriguez — Sun, 17 Oct 2021 19:58:48 +0000

Welcome back to the next chapter of { Zero to Helm }. A blog-series in which I create "byte"-sized pieces of content aimed at mastering Kubernetes.

In this section, we are going to look to take a surface dive into Kubernetes by inspecting its architecture. Whether you are a fan of minimal version of k8s like k3s or robust cloud-specific versions like EKS, AKS, or GKE - the information explained here applies to all of them.

So lets start to breakdown this architecture:

Nodes

Kubernetes at its essence is a collection of servers that work together to achieve a single purpose; run your workloads. These servers are referenced as "Nodes", and their responsibility is to run your workloads by placing containers into Pods.

Since Nodes are servers, they can be anything from physical machines to even virtual ones! But each node is managed by something called a "control plane". Which is a collection of essential services that run on the node in order for Pods to be created, and managed by the node. Those services are the kubelet, a container runtime, and the kube-proxy.

Once a node is connected to a Kubernetes master node, it will begin to report vital information like:

CPU/Memory availability
Info about the host/kernel
Heartbeats

All of this information help the control plane make decisions about where workloads get scheduled/assigned.

Control Plane Communication

When we say “Control Plane” - we are speaking about the essential services that run on the master nodes of your Kubernetes Cluster. Collectively, these services help Kubernetes perform most of its magic for being easily extensible, accessible, and able to run workloads efficiently.

Now - those magical services are:

kube-apiserver

Exposes the Kubernetes API that is used by worker nodes, kubectl, and other services to interact with the entire k8s control plane.
etcd*

This service can differ based on your k8s flavour but a simple highly available key/value store is used in k8s to track cluster states such as pod information, workload assignments, etc.
kube-scheduler

Watches for new pods without a node assignment and selects one for them. Various factors are taken into account but the assignment is based on the pod’s manifest and key scheduling information collected from the cluster’s available nodes.
kube-controller-manager
Runs controller processes like:
- Node Controller => Monitors cluster nodes
- Jobs Controller => Monitors Pods that run one-off tasks
cloud-controller-manager*
- This one is a bit of a special one because not every flavour of k8s will include it. But this service typically handles cloud specific control logic like identity permission controls, network rules, and anything else your cloud-hosting provider specifically requires for their flavour of k8s.

More info: Kubernetes Components | Kubernetes

And that’s it! It really is that simple at a high level. There is a lot more additional terminology like Services, Ingress, and Namespaces - but all in good time! The point here is not to overwhelm but rather to encourage individual discovery.

There are a lot of helpful links throughout here and I highly encourage you to dive deeper and investigate what other flavours of k8s look like! How do they differ? What’s similar…?!

As always, feel free to give feedback and let me know about your journey with Kubernetes! Until next time, happy coding!

{ Zero to Helm }: Part 1 - History

Stephen Rodriguez — Sun, 05 Sep 2021 10:39:29 +0000

Welcome to the first entry of the { Zero to Helm } series! On this weeks byte, we will be going over a brief recap of some of the greatest moments Kubernetes has had from the past two decades. Yes, you read that right - the origins of Kubernetes stems back as far as the year 2003 with a system known as Borg! But don't fret, we will make sure to keep you up to speed with some recent history too and maybe some insight into what might come next!

🌱 2003 - The Early Days of Cluster Management

Back in 2003, Google had introduced a new cluster management tool called Borg. Now, don't worry, we aren't talking about these kind of Borgs from Star Trek (although I wouldn't put it past them to use it as a pun). Borg was a privatized tool that only Google had used to manage the kinds of things we use Kubernetes for today.
During these early days, Borg was used for everything from running websites, stateful servers, and even complete application infrastructures like BigTable, and MapReduce. Although Google had Borg, other cluster-management tools did eventually exist for others to use. Some of these were Apache Mesos, Microsoft Cluster Server, IBM's Tivoli System, and eventually Docker Swarm. Despite the massive adoption's these other tools had, Borg still had a certain finesse to it that other's did not possess. Some of those aspects still exist to this day in Kubernetes. Ideas such as Pods, Services, and Labels still remain some of the most core functional components of any Kubernetes cluster.

Highlights

2003: Borg is born

🍁 2014 - The Birth of Kubernetes

Fast-forward a decade and Google introduced its open-sourced version of Borg on Github, and thus Kubernetes was born. It's founding fathers were Joe Beda, Brendan Burns, and Craig McLuckie - some of the brightest minds that built Google's Cloud Computing Services. The name is a bit of a nod to the Borg name. Internally, Kubernetes was code-named as Project 7; a nod to an ex-borg character from Star-Trek. The logo (with seven spokes) was also a nod to it's former codename and predecessor.

At the time, Github was (and still-is) a staple in any project as a well known source for famous open-source projects like React, Typescript, and even Linux. It was only natural that Google would do the same as they have with other projects (i.e. Angular), and early adoption was quick!
It didn't take very long until other tech giants joined suit in support for this new endeavor; e.x. Microsoft & IBM.

Highlights

2014: Kubernetes is born (Codename: Project 7)

Originally written in C++
Eventually rewritten using Go

🌳 2015–2017 - Road towards a Stable "Foundation"

The early days of Kubernetes were a bit rocky. As with any new open-sourced project, the eyes of scrutiny were placed tightly on Kubernetes now that the world could contribute on Github. It wasn't until mid-2015 that the first major release of Kubernetes was seen. This major release also formed a launchpad for its explosive adoption. Any first major release of an open-source project is a cornerstone for success and a cause for celebration!

In junction with the 1.0 release, Google doubled down on it's commitment to the open-source community by collaborating with the Linux Foundation to form a new foundation entirely focused on the key tenants of Kubernetes; containerization & microservice architecture. That organization is the Cloud Native Computing Foundation (CNCF).

Over the next two years, Kubernetes (in tandem with the foundation) had moments of massive global growth as others aimed to capitalize on it's promise. This is when we saw the birth of tools like Helm (package management on k8s), Minikube (local k8s), and Kops (production k8s). We even saw companies like Pokemon GO! share their own success story. All of this brought about a growing sense of confidence that Kubernetes was bound to become a staple in every office's long-term plan for cloud infrastructure.

By March 2017, Kubernetes 1.6 was released and marked by Google as a stabilization of the platform. From this point onward, the world felt much safer with switching over and running their own Kubernetes setups in production. Soon after, companies like Github announced their own case study on how they made the switch to Kubernetes in production.
Even other cloud platforms like Amazon & Microsoft announced new service offerings for Kubernetes in the cloud. It didn't take long for anyone to see that adoption should be much higher on their backlog of upcoming projects.

Highlights:

2015: Kubernetes v1.0
2015: Google/Amazon/Microsoft offer Kubernetes in the cloud Linux + Google open the CNCF
All future releases of Kubernetes is managed by the CNCF
Kubecon begins as the official Kubernetes conference
CNCF offers incubation program
Kubernetes is the first to graduate from incubation

👥 2017–2021: "Community-driven-Growth"

As we fast forward through the next couple of years, Kubernetes adoption was loud & wild. FOMO energies spiked, and every R&D department in the world had at least one person singing it's song. Everything from Serverless (i.e. Knative) to Stateful architectures were being redesigned for this new world. International forums like Kubecon (an annual conference managed by the CNCF about K8s) provided a meeting ground for local-communities to share & engage with one another on their own successes/failures, and made it easy for newcomers to feel welcomed to this new world of Kubernetes.

In addition to all of this community engagement, the CNCF also formed an incubation program for up & coming cloud-based projects. This program served as a kind of launchpad for exciting new projects that could further empower existing Kubernetes users. This program became a recipe for boosting the return on early adopter investment as they were the one's who were able to capitalize on projects that graduate from the program. One example of this was the birth of AWS App Mesh & it's use of the CNCF graduated project - Envoy Proxy.

Altogether, both the forums & incubation program formed a recipe for success! From here on out, it has become a matter of keeping the community engaged & informed while also going back to Github for open-source contributions. Kubernetes is definitively the king of the market today for cluster management. All major cloud providers offer their own hosting services, and the barrier for entry has never been lower for both small & large companies.

🔮 The Future of Kubernetes

At this point, it really is up to the community. Github & the CNCF serves as public forums for the world to engage with each other on patching & improving the Kubernetes platform. In addition, the CNCF's incubation program provides an avenue for enthusiasts to gain support for their ambitious ideas that can empower the world of Kubernetes.

Altogether, it's hard to say where things will go in terms of new ground. However, what we will see if the adoption of Kubernetes for pretty much every market sector out there. Even IoT has seen the power of Kubernetes with the release of K3S by Rancher Labs. Frankly, the limits just don't exist, and the possibilities are endless.

Next Week on { Zero to Helm }…

-Part 2: Kubernetes Architecture-

Next week we will start diving into the Kubernetes ecosystem. Pointing out the various parts of the system & their responsibilities before we begin to dive deeper & getting our hands dirty via code.

If you are interested in following my series on all-things Kubernetes, consider following me on Dev.To.

{ Zero to Helm }: Introduction

Stephen Rodriguez — Sun, 05 Sep 2021 10:24:03 +0000

It doesn't take a lot of digging around on the internet to see that Kubernetes has become a staple in any modern developer's environment. Maybe you heard about it from a conference? Maybe your co-worker who can't stop bragging about it? - or perhaps you have had it on your personal wishlist of next things to learn? Well, this is me telling you that now is the time. Now is your opportunity to finally get started and learn everything you need to know about Kubernetes & Helm. Now is your chance to go from { Zero to Helm } !

Who am I?

My name is Stephen Rodriguez and I have been working as a Full-Stack Engineer & as a DevOps engineer for the better part of 8 years. I have both a B.S. & M.S. degree in Computer Science and have had my research published by in three different IEEE journals.

Over the past years, I have worked with both small & large corporations to help develop their IPs by constructing DevOps pipelines & procedures to support hosting their services in the cloud and to meet the demands of a global audience. Pre-COVID, I gave talks in my hometown and offered lectures at my place of work to help R&D teams catch up with all the nuances of Cloud Services, Kubernetes, and DevOps Best Practices.
Today, I work at Sharkmob. A gaming company based in Malmo, Sweden. Here, I develop our DevOps platform utilizing Kubernetes via our Live Operations division.
Shameless plug to my LinkedIn.

What is { Zero to Helm } ?

{ Zero to Helm } is a progressive series of digestible "bytes" that will help you pick up on all the nuances of Kubernetes & Helm. Each byte is meant to be consumed in less than an hour and provide a wealth of knowledge about one specific aspect of Kubernetes. Each byte will build upon its predecessors and soon enough - you too will be mastering the helm of Kubernetes!

Co-worker: What's for lunch today?
Me: Hm, the new { Zero to Helm } article is out, want a byte?

What is on the roadmap?

Brief History of Kubernetes
Development Setup (circa. 2021)
Local Kubernetes Setup
Configuration as Code (YAML)
Pods & DaemonSet
Services
Ingress
Deployments & ReplicaSets
Deployment Strategy & Rollbacks
( and so much more… )

There is a lot to learn about Kubernetes, but we have to begin somewhere. So why not at the very beginning? We will begin with some origin stories about the early days of Kubernetes and what Container Orchestration is as a whole. This smaller set of bytes ("bits") will highlight Google's Borg project and other similar competitors that floated around the ether before Kubernetes. Afterwards, we will help prepare your development environment with the right tools & resources you will need in order to master Kubernetes like a Pro. Once set up, each byte afterwards will focus on introducing you to Kubernetes and expanding your knowledge of the various resources it provides. This includes everything from running your application within a Pod to building a production-grade DevOps pipeline for continuous deployment!

What's next?

If I haven't lost you yet - consider subscribing to my posts to get alerts when the next article is released! I am aiming to push these out on a weekly basis - and as always, I am open to feedback/suggestions on what you want to learn!

{ Zero to Helm } - The Mission - Part 1

Stephen Rodriguez — Mon, 11 Jan 2021 20:24:29 +0000