Flyte School: Kubernetes for ML and Data - An Introduction
As Kubernetes adoption continues to soar, data scientists are presented with an opportunity to transform their workflow and drive innovation with enhanced scalability, agility, and resource optimization. Whether you're a seasoned data scientist or just starting out, this talk will equip you with the foundational knowledge needed to harness the power of Kubernetes and take your data science projects to new heights.
The Agenda Includes
- Why Kubernetes?
- Kubernetes architecture: a logical overview
- Key constructs: Pods, Services, Deployments and more
- Data management
Learning Goals
- Understand the building blocks of a Kubernetes system, their relationships and hierarchy
- Learn how to leverage Kubernetes resources to transform the development lifecycle
- Deploy a simple workload and see how Kubernetes handles data persistency
Prep Material
- Non-heavily technical intro material: The Illustrated Children’s Guide to Kubernetes | Cloud Native Computing Foundation
- To spin up a cluster use either Play with Kubernetes or the local minikube package