Kubernetes KUBERNETES
EKS
K8s
GKE AWS EKS Kubernetes
"Kubernetes
- Kubernetes 1.33: Job’s SuccessPolicy Goes GA 15MAY2025 (Kubernetes) => Job Success Policy (GA) allows tailoring of success criteria for batch runs containing failures (useful in AI/ML & HPC scenarios).
- Kubernetes v1.33: Updates to Container Lifecycle/ 14MAY2025 (DigitalOcean) => Sleep & Stop Signals in Container Lifecycle…
- Kubernetes v1.33: Job’s Backoff Limit Per Index Goes GA 13MAY2025 (Google) => Backoff Limit Per Index feature for fine-tuning batch error handling…
- Kubernetes v1.33: Image Pull Policy the way you always thought it worked!/ 12MAY2025 (Microsoft) => Fix 10 years in the making (correcting auth behavour)…
- Kubernetes v1.33: Streaming List responses 09MAY2025 (Google) => Tuning list memory consumption for large datasets…
- Kubernetes 1.33: Volume Populators Graduate to GA 08MAY2025 (Google) => volume populators using custom resources are now generally available (GA)!
- Kubernetes v1.33: From Secrets to Service Accounts: Kubernetes Image Pulls Evolved 07MAY2025 (Micosoft) => Service Account Tokens as creds for pulling images…
- Kubernetes v1.33: Fine-grained SupplementalGroups Control Graduates to Beta 06MAY2025 (LY Corp) => securing supplementary groups in security context
- Kubernetes v1.33: Prevent PersistentVolume Leaks When Deleting out of Order graduates to GA 05MAY2025 (Broadcom) => Fixing resource reclaim issues when deleting a PV before a PVC…
CNCF Cloud Native AI (CNAI) Paper is out! (KubeCon24)
Refer to the following for link to download the CNAI working group report (Mar 2024, CNCF presented @KubeCon24 Paris)
CNCF AI Working Group – Cloud Native Artificial Intelligence Paper
Blogs
Container Orchestration is Everywhere...
While the use of Docker Compose is often viable in development contexts, CNCF Kubernetes is the contemporary go-to technology for production-grade secure container orchestration at scale, able to provisioning resilient, fault-tolerant workloads (usually) in multi-tenanted contexts.
Kubernetes (K8S) deployment options range from …
- performing a ‘fully-manual’ cluster install (kubeadm) on local infrastructure / on-prem.
- use of minikube to spin-up a quick-&-dirty lightweight local (single-node) cluster for development purposes (only).
- full-manual installation of cluster on virtual servers from your Cloud-Vendor of choice (IaaS).
- use of AWS Elastic Kubernetes Service (EKS) where Amazon as Cloud-Vendor provisions K8s cluster (PaaS).
- use of GCP Google Kubernetes Engine (GKE) Service leveraging Google as Cloud-Vendor managing K8s Cluster provisioning (PaaS).
Amazon EKS
Amazon offers the managed Elastic Kubernetes Service (EKS) which provisions an AWS-hosted Kubernetes Control Plane.
NEW Certified Kubernetes Administrator (CKA) Guide
This is the guide to use to master the current CNCF Kubernetes CKA exam. Learn & certify on the Kubernetes skills required to administer and operate Kubernetes clusters at the professional level, and advance your AWS (EKS) career !
Should I do the AWS EKS Knowledge Badge??
While AWS Certifications span the width & breadth of capabilities on the AWS Cloud, there is currently no AWS Platform/Orchestration certification based on use of Kubernetes. However, platform engineers looking to certify K8s skills on AWS should consider gaining the AWS EKS Knowledge Badge, available for free via AWS SkillsBuilder.
While not as challenging as the CNCF (CKA,CKS, CKAD etc) accreditations, this AWS EKS course contains a series of technical EKS overviews, providing technical insights into contemporary aspects of the AWS EKS service. The course modules touch on the following notable contemporary EKS patterns & practices (including use of EKS Add-Ons & Third-Party Packages).
* AWS Distro for OpenTelemetry (ADOT) Collector (for example for X-Ray traces).
* FluentBit (for CloudWatch Logging)
* AWS Load Balancer Controller (for NLB load balancing & ALB provision for Ingress).
* Inspector Image Scanning (Advanced ECR Scanning).
* Flux versus ArgoCD (discussion).
* Security Groups for Pods (AWS VPC CNI)
* EKS Auto Mode (AWS managed resource optimisations to reduce costs).
* Kubecost (fine-grained cost reporting at K8s namespace, label & pod level).
Course contains a free assessment exam requiring an 80%+ result to gain the EKS Badge accreditation. This course is not the place to start for someone learning K8s. This course is best suited to platform engineers with intermediate-level hands-on K8s experience who will have the background knowledge to ‘fill in’ the gaps in this course around how K8s works. For experienced K8s engineers, investing the ~10 hours required to complete the course will provision useful technical EKS insights around the AWS management of the K8s Control Plane. And of course the corresponding AWS EKS Knowledge Badge accreditation!
Design Choices for your K8s Cluster
1) SSH vs SSL – Which Scales Better?
↳ https://lnkd.in/eZM8iBfa
2) Conventional Vs Kubernetes CI/CD Pipelines
↳ https://lnkd.in/ePmpBy53
3) Dockerfile vs Docker Compose: What You Should Know
↳ https://lnkd.in/eUX42Xcw
4) Kubernetes Operator vs Helm – Which One to Choose?
↳ https://lnkd.in/efU-YU6Y
5) AWS Internet Gateway vs NAT Gateway – Which One to Choose?
↳ https://lnkd.in/ebKG-p84
6) Kubernetes Ingress Vs Gateway API
↳ https://lnkd.in/epMgWD5g
7) GitOps Push vs Pull Architecture – Which One to Choose?
↳ https://lnkd.in/eTpQkGPC
8) CloudFront Signed URL vs S3 Pre Signed URL – When to Use What?
↳ https://lnkd.in/epD7g2KB
9) HPA vs VPA vs KEDA – Kubernetes Autoscaling
↳ https://lnkd.in/exehKJ_F