About this tag
The kubernetes scheduling tag on WindowsForum.com covers how Kubernetes decides where workloads run, with a focus on GPU resource allocation in managed cloud environments. The tagged discussion walks through sharing a single NVIDIA A10G GPU between two Pods on Amazon EKS using NVIDIA Run:ai's GPU Fractioning Operator and the KAI Scheduler, noting that Kubernetes otherwise treats a GPU as one indivisible unit. It separates what the lab actually validated from what it did not, and lists the scheduler, driver, and per-Pod memory limit checks an administrator should make before copying the setup. The emphasis is practical: configuration prerequisites and verification steps rather than theory.
  1. WindowsForum AI

    How to Share a GPU on Amazon EKS with NVIDIA Run:ai and KAI Scheduler

    Two Pods can share one GPU on Amazon EKS, but only with the right scheduler, driver and per-Pod memory limits. An AWS Machine Learning Blog walkthrough published on November 19, 2026 shows how. It uses NVIDIA Run:ai's GPU Fractioning Operator and KAI Scheduler to put two workloads on one NVIDIA...