Learn from production

• An 8-minute outage from a dead NLB and a JVM that cached DNS forever

  claire nguyen

  This case study shows how a JVM DNS cache pinned one service to dead AWS Network Load Balancer IPs during migration, causing an outage fixed with a 30-second TTL.

• Load Testing Apollo Router

  Yurii Rochniak

  This case study shows how Preply tested Apollo Router with a Rust mock service and Vegeta, found a safe range of 1,200–1,500 requests per second, and chose CPU-based autoscaling.

• Your KV Cache Transfer Is 100x Slower Than It Should Be

  Federico Iezzi

  This case study finds a 100x KV cache transfer slowdown in vLLM on GKE, shows how NixlPushConnector restores about 49 GB/s, and measures the latency tradeoff of disaggregated LLM serving.

• OpenShift Upgrade on IBM Cloud with Portworx Replication Enabled: Lessons from a Production Upgrade

  Akshay A Jain

  This case study shares how a team upgraded a production OpenShift cluster on IBM Cloud with Portworx replication in place, and which compatibility and worker-replacement checks kept it from turning into an outage.

More case studies →

Matching jobs

1. • DFIR Expert with Nebius

   • Salary: $0 to $517.18K a year

   • Location: remote from

   • Tech stack: Kubernetes, Powershell, Python, SQL

2. • Data Engineer with Mobiik

   • Salary: $112.5K to $220K a year

   • Location: fully remote

   • Tech stack: Kubernetes, Azure, Docker, SQL

3. • Data Engineer with MyFitnessPal

   • Salary: $160K to $220K a year

   • Location: remote from

   • Tech stack: Kubernetes, AWS, Docker, SQL, Python, Scala, Snowflake, Kafka, Airflow

4. • Data Engineer with Valarian Technologies Limited

   • Salary: US$112.5k to US$291.5k a year

   • Location: based in the office (and remote from home) in London, GB

   • Tech stack: Kubernetes, On-premise, OpenShift, Helm, ArgoCD, Python, Shell, SQL, PostgreSQL, Airflow

5. • DevOps Engineer with OKX

   • Salary: $214.67K to $322K a year

   • Location: based in the office in San Jose, CA, USA

   • Tech stack: Kubernetes, AWS, Go, Python, Shell, SQL, Redis, PostgreSQL, MySQL, Terraform

Discover more Kubernetes jobs on Kube Careers →

Build something

• Migrating from Ingress NGINX to Gateway API on EKS

  Douglas Alves

  This tutorial shows how to move an EKS cluster from the retired ingress-nginx controller to Gateway API, covering the AWS Load Balancer Controller CRDs and a zero-downtime traffic shift.

• Building an InfluxDB OSS Provider with Crossplane and Upbound — My Hands-On Platform Engineering Lab

  Subasukumar

  This tutorial walks through building a Crossplane platform API on a local kind cluster, modelling an InfluxDB provider with XRDs and compositions, and debugging the composition when it fails.

• Migrating from AWS VPC CNI & Kube-Proxy to Cilium on EKS

  Pavankumarn

  This tutorial teaches how to migrate EKS from AWS VPC CNI and kube-proxy to Cilium using blue/green nodes, WireGuard encryption, and Hubble for observability.

• Building a 100% Local RAG System on Kubernetes

  Ahmed Nafies

  This tutorial shows how to run a fully local RAG system on Kubernetes with PostgreSQL and pgvector, local embeddings, and a CPU-based Qwen model without API keys.

More tutorials →

Call for Papers closing soon

1. 2

   days

   Open Source Conference Kolkata

   The Call For Paper is open until 5 November 2026 at GMT-5. More info →

   • Location: Kolkata, IN

   • In-person conference organized by Open Source Con India.

   • The conference starts on the 5 December 2026.

   • Apply here

2. 20

   days

   Big Data Europe

   The Call For Paper is open until 23 November 2026 at GMT-5. More info →

   • Location: Vilnius, LT

   • In-person conference organized by DATA MINER.

   • The conference starts on the 27 November 2026.

   • Apply here

3. 28

   days

   ContainerDays & AI Context London 2027

   The Call For Paper is open until 1 December 2026 at GMT-5. More info →

   • Location: London, GB

   • In-person conference organized by Dockland Media Group.

   • The conference starts on the 11 February 2027.

   • Apply here

4. 39

   days

   APE 2026

   The Call For Paper is open until 12 December 2026 at GMT-5. More info →

   • Location: Apenheul, NL

   • In-person conference organized by Stichting Meetups.

   • The conference starts on the 11 December 2026.

   • Apply here

5. 45

   days

   Dawn Conferences 2027

   The Call For Paper is open until 18 December 2026 at GMT-5. More info →

   • Location: Amsterdam, NL

   • In-person conference organized by Dawn Technology..

   • The conference starts on the 19 March 2027.

   • Apply here

6. 58

   days

   Devopsdays Raleigh

   The Call For Paper is open until 31 December 2026 at GMT-5. More info →

   • Location: Raleigh, NC, USA

   • In-person conference organized by Devopsdays.

   • The conference starts on the 27 April 2027.

   • Apply here

7. 124

   days

   Devopsdays Geneva

   The Call For Paper is open until 7 March 2027 at GMT-5. More info →

   • Location: Geneva, CH

   • In-person conference organized by Devopsdays.

   • The conference starts on the 25 May 2027.

   • Apply here

More articles

• I benchmarked 5 Languages in Kubernetes and here is what your stack actually costs

  Ashok Vishwakarma

  This article benchmarks Go, Node.js, C#, Java and Python on the same Kubernetes cluster and shows how image size and startup time drive scaling speed and turn language choice into a monthly bill.

• I Don’t Trust etcd and Neither Should You

  Har8

  This article explains why Kubernetes keeps all its truth in etcd, and how the Raft algorithm gets several servers to agree on one version of reality before your apply returns.

• We forked Apache Stateful Functions for Flink 2.x — here's why

  Oleksandr Kazimirov

  This article explains why a team forked Apache Stateful Functions to keep it running on Flink 2.x and Java 21, what they changed in the connectors, build and release pipeline, and when you should use the fork.

• I Built an LLM Platform Without Burning Cash on GPUs

  Harshitha Anuganti

  This article shows how an LLM inference platform was built on EKS with Terraform, GitOps and monitoring using a cheap CPU stub first, so the GPU only arrived once the whole platform worked.

Even more articles →