Kubernetes bitesize series built for busy people. No more than 2 min read. Read it while waiting for the coffee to brew.
Kubernetes bitesize blog posts attracted a lot of curiosity and questions. Here are Q&A that you can greatly benefit from:
Question by Taragurung:
Hi Taragurung, you are absolutely correct. In the end, the system is bounded by the number of nodes available and yes there is latency in how long it takes to the machine to spin a new pod assuming a node is available. What exists in k8s is a load balancer that routes the messages for various pods replications. This is not enough, what can help is having some message buffer/messaging system like Kafka to create back pressure on the system. or to decide to drop some of the traffic - this is, of course, depends on the product use case.
Question by wintvels:
Hi! Thank you for asking. I will do my best to explain.
Let's start from the beginning:
what is a cluster? a cluster is a number of computers ( sometimes we call them machines or nodes - they can be physical or virtual) that work together to fulfill a target.
when we say Kubernetes cluster - we mean a group of machines that has Kubernetes software installed on.
Each of the machines has a role, they either act as masters or workers.
In a simple state, one machine is acting as Master and the rest acts as Workers - meaning they get tasks and answers the Master while the master is in charge of distributing the tasks in general.
The workers always report to the master on their state and the master is constantly monitoring the workers to validate that everything is running as expected.
Good place for basic core concepts is this one
another good place is this one: https://kubernetes.io/docs/concepts/ but this one is deep dive and can be a bit confusing to start with so it's better to start with the first link and then go to this one. I hope this helps!
Thank you, everyone, who wrote me in private and public and shared my blog posts with the world 🌎 🌍 🌏
If you have more questions and ideas for me, please reach out!
I'm available here and on Twitter.
Missed the series and want to learn about K8s, here are links for you to get started fast: