kubernetes-sigs/dra-driver-nvidia-gpu

آخر نشاط مصدر مسجل: ٢٨ أغسطس ٢٠٢٦ في ١٨:٤٢

آخر تحديث لفهرس SkillsMP: ١٢ نوفمبر ٢٠٢٦ في ١٢:٥٢

skills مجمعة: ٩

نجوم GitHub: ٧٢٨
تفرعات GitHub: ٢١٧

التثبيت بمساعدة مساعد ذكي

انسخ هذا الطلب إلى المساعد الذكي الذي تستخدمه.

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/kubernetes-sigs/dra-driver-nvidia-gpu.

ملف README للمستودع

مصدر ملف README للمستودع: README.md

قراءة الملف الأصلي

عرض ملف README كاملًا

# DRA Driver for NVIDIA GPUs Enables * flexible and powerful allocation and dynamic reconfiguration of GPUs as well as * allocation of ComputeDomains for robust and secure Multi-Node NVLink. For Kubernetes 1.32 or newer, with Dynamic Resource Allocation (DRA) [enabled](https://kubernetes.io/docs/concepts/scheduling-eviction/dynamic-resource-allocation/#enabling-dynamic-resource-allocation). ## Overview DRA is a novel concept in Kubernetes for flexibly requesting, configuring, and sharing specialized devices like GPUs. To learn more about DRA in general, good starting points are: [Kubernetes docs](https://kubernetes.io/docs/concepts/scheduling-eviction/dynamic-resource-allocation/), [GKE docs](https://cloud.google.com/kubernetes-engine/docs/concepts/about-dynamic-resource-allocation), [Kubernetes blog](https://kubernetes.io/blog/2025/05/01/kubernetes-v1-33-dra-updates/). Most importantly, DRA puts resource configuration and scheduling in the hands of 3rd-party vendors. The DRA Driver for NVIDIA GPUs manages two types of resources: **GPUs** and **ComputeDomains**. Correspondingly, it contains two DRA kubelet plugins: [gpu-kubelet-plugin](https://github.com/kubernetes-sigs/dra-driver-nvidia-gpu/tree/main/cmd/gpu-kubelet-plugin), [compute-domain-kubelet-plugin](https://github.com/kubernetes-sigs/dra-driver-nvidia-gpu/tree/main/cmd/compute-domain-kubelet-plugin). Upon driver installation, each of these two parts can be enabled or disabled separately. The two sections below provide a brief overview for each of these two parts of this DRA driver. ### `ComputeDomain`s An abstraction for robust and secure Multi-Node NVLink (MNNVL). Officially supported. An individual `ComputeDomain` (CD) guarantees MNNVL-reachability between pods that are _in_ the CD, and secure isolation from other pods that are _not in_ the CD. In terms of placement, a CD follows the workload. In terms of lifetime, a CD is ephemeral: its lifetime is bound to the lifetime of the consuming workload. For more background on how `ComputeDomain`s facilitate orchestrating MNNVL workloads on Kubernetes (and on NVIDIA GB200 systems in particular), see [this](https://docs.google.com/document/d/1PrdDofsPFVJuZvcv-vtlI9n2eAh-YVf_fRQLIVmDwVY/edit?tab=t.0#heading=h.qkogm924v5so) doc and [this](https://docs.google.com/presentation/d/1Xupr8IZVAjs5bNFKJnYaK0LE7QWETnJjkz6KOfLu87E/edit?pli=1&slide=id.g28ac369118f_0_1647#slide=id.g28ac369118f_0_1647) slide deck. For an outlook and specific plans for improvements, please refer to the [project releases](https://github.com/kubernetes-sigs/dra-driver-nvidia-gpu/releases). If you've heard about IMEX: this DRA driver orchestrates IMEX primitives (daemons, domains, channels) under the hood. ### `GPU`s The GPU allocation side of this DRA driver [will enable powerful features](https://docs.google.com/document/d/1BNWqgx_SmZDi-va_V31v3DnuVwYnF2EmN7D-O_fB6Oo) (such as dynamic allocation of MIG devices). To learn about what we're planning to build, please have a look at the [project releases](https://github.com/kubernetes-sigs/dra-driver-nvidia-gpu/releases). While some GPU allocation features can be tried out, they are not yet officially supported. Hence, the GPU kubelet plugin is currently disabled by default in the Helm chart installation. For exploration and demonstration purposes, see the "demo" section below, and also browse the `demo/specs/quickstart` directory in this repository. ## Installation Configuration and installation instructions are in the [documentation site](https://dra-driver-nvidia-gpu.sigs.k8s.io/docs/install/) ([source](site/content/docs/install.md)). ## A (kind) demo Below, we demonstrate a basic use case: sharing a single GPU across two containers running in the same Kubernetes pod. **Step 1: install dependencies** Running this demo requires * kind (follow the official [installation docs](https://kind.sigs.k8s.io/docs/user/quick-start/#installation)) * NVIDIA Container Toolkit & Runtime (follow a [previous version](https://github.com/kubernetes-sigs/dra-driver-nvidia-gpu/blob/5a4717f1ea613ad47bafccb467582bf2425f20f1/README.md#demo) of this readme for setup instructions) **Step 2: create kind cluster with the DRA driver installed** Start by cloning this repository, and `cd`in into it: ```console git clone https://github.com/kubernetes-sigs/dra-driver-nvidia-gpu.git cd dra-driver-nvidia-gpu ``` Next up, build this driver's container image and create a kind-based Kubernetes cluster: ```console export KIND_CLUSTER_NAME="kind-dra-1" ./demo/clusters/kind/build-dra-driver-gpu.sh ./demo/clusters/kind/create-cluster.sh ``` Now you can install the DRA driver's Helm chart into the Kubernetes cluster: ```console ./demo/clusters/kind/install-dra-driver-gpu.sh ``` **Step 3: run workload** Submit workload: ```console kubectl apply -f ./demo/specs/quickstart/v1/gpu-test2.yaml ``` Container log inspection then indeed reveals that both containers operate on the same GPU device: ```bash $ kubectl logs pod -n gpu-test2 --all-containers --prefix [pod/pod/ctr0] GPU 0: NVIDIA A100-SXM4-40GB (UUID: GPU-4404041a-04cf-1ccf-9e70-f139a9b1e23c) [pod/pod/ctr1] GPU 0: NVIDIA A100-SXM4-40GB (UUID: GPU-4404041a-04cf-1ccf-9e70-f139a9b1e23c) ``` ## Contributing Contributions require a Developer Certificate of Origin (DCO, see [CONTRIBUTING.md](https://github.com/kubernetes-sigs/dra-driver-nvidia-gpu/blob/main/CONTRIBUTING.md)). ## Support Please open an issue on the GitHub project for questions and for reporting problems. Your feedback is appreciated!

جارٍ تحميل ملف README للمستودع…

Skills في هذا المستودع

مطوّرو البرمجيات محللو ضمان جودة البرمجيات والمختبرون

٢ فئات مهنية · 100% مصنفة