精选8805:21MarkTechPost（RSS）产品发布/更新

微软开源TauGrid：Kubernetes原生GPU AI工作负载栈

Microsoft Open-Sources TauGrid: A Kubernetes-Native Stack for GPU AI Workloads

推荐理由

做MLOps和K8s基础设施的同学必看，TauGrid把原本需要手动拼接的GPU调度、Ray编排和监控组件打包成一键Helm部署，且默认无遥测，非常适合私有化部署场景，建议直接去GitHub看文档评估集成成本。

Platform teams running AI on Kubernetes rarely run one thing. They run a queueing system, a distributed runtime, GPU node health checks, dashboards, and a layer of submission scripts holding all of it together. The Azure Kubernetes Service engineering team open-sourced TauGrid, which collapses that assembly job into a single Helm install.

在 Kubernetes 上运行 AI 的平台团队很少只运行单一组件。他们通常会部署队列系统、分布式运行时、GPU 节点健康检查、仪表盘，以及将所有这些组件整合在一起的提交脚本层。Azure Kubernetes Service (AKS) 工程团队开源了 TauGrid，它将这一系列组装工作简化为一次 Helm 安装。

Is it deployable? Yes, TauGrid is MIT licensed, with container images and Helm charts published as public OCI artifacts on Microsoft Container Registry. Prerequisites are a Kubernetes 1.30+ cluster with GPU nodes, kubectl, and Helm 3.0 or later.

是否可部署？是的，TauGrid 采用 MIT 许可证，容器镜像和 Helm Chart 已作为公共 OCI 制品发布在 Microsoft Container Registry (MCR) 上。前置条件包括：带有 GPU 节点的 Kubernetes 1.30+ 集群、kubectl 工具以及 Helm 3.0 或更高版本。

What is TauGrid

什么是 TauGrid

TauGrid is a self-hosted platform for running AI workloads on Kubernetes. It combines five things that platform teams usually integrate by hand: the tau CLI, workload queueing and admission through Kueue, Ray cluster orchestration through KubeRay, node-level GPU health monitoring, and cluster and workload observability.

TauGrid 是一个自托管平台，用于在 Kubernetes 上运行 AI 工作负载。它整合了平台团队通常手动集成的五个部分：tau CLI、通过 Kueue 实现的工作负载排队与准入、通过 KubeRay 实现的 Ray 集群编排、节点级 GPU 健康监控，以及集群和工作负载的可观测性。

The split of responsibility is the design point. Platform teams own workspaces, queues, compute profiles, storage, identity, and observability. Researchers work from a repository and the CLI, and submit workloads without configuring Kubernetes directly. The codebase is written primarily in Go.

职责划分是其设计核心。平台团队负责管理工作区、队列、计算配置文件、存储、身份认证和可观测性。研究人员从代码库和 CLI 进行操作，直接提交工作负载而无需配置 Kubernetes。该代码库主要使用 Go 语言编写。

How a job moves through it

任务在其间的流转过程

A workload is described in a tau.yaml file. The GPU training example published by Microsoft runs a PyTorch job on a single A100:

工作负载在 tau.yaml 文件中进行描述。微软发布的 GPU 训练示例在单个 A100 GPU 上运行一个 PyTorch 任务：

代码 · 14 行

schema_version: 1
name: aks-gpu-quickstart
run:
  entrypoint: train.py
  workload_kind: rayjob
compute:
  gpus: 1
  workers: 1
  cpus: 16
  memory: 64Gi
runtime:
  image: mcr.microsoft.com/aks/ai-runtime/ray:py3.12-ray2.56.0-cuda13.0
  pip:
    - torch>=2.4.0

代码 · 14 行

schema_version: 1
name: aks-gpu-quickstart
run:
  entrypoint: train.py
  workload_kind: rayjob
compute:
  gpus: 1
  workers: 1
  cpus: 16
  memory: 64Gi
runtime:
  image: mcr.microsoft.com/aks/ai-runtime/ray:py3.12-ray2.56.0-cuda13.0
  pip:
    - torch>=2.4.0

On tau run, TauGrid resolves platform policy, renders a Kubernetes Job or a KubeRay RayJob, and submits it through Kueue. The six stages Microsoft documents are submission, queueing, execution, monitoring, recovery, and evidence. Recovery covers retry, resume from checkpoint, and failure diagnosis. Evidence records capture workload metadata, configuration, logs, metrics, checkpoints, and execution history, which is what makes a run reproducible and auditable later.

执行 `tau run` 时，TauGrid 解析平台策略，渲染 Kubernetes Job 或 KubeRay RayJob，并通过 Kueue 提交。微软文档中描述的六个阶段为：提交、排队、执行、监控、恢复和证据记录。恢复阶段涵盖重试、从检查点恢复以及故障诊断。证据记录捕获工作负载元数据、配置、日志、指标、检查点和执行历史，这使得后续的运行可复现且可审计。

When several teams share a cluster, their jobs land in a shared Kueue ClusterQueue. Kueue admits each one on quota and priority, and Kubernetes places it on healthy GPUs.

当多个团队共享一个集群时，他们的任务会进入共享的 Kueue ClusterQueue。Kueue 根据配额和优先级接纳每个任务，Kubernetes 则将其调度到健康的 GPU 上。

Interactive explainer

交互式解释器

Install footprint

安装占用空间

Installation is a Helm chart pulled straight from MCR:

安装过程是从 MCR 直接拉取一个 Helm Chart：

代码 · 5 行

helm install taugrid \
  oci://mcr.microsoft.com/aks/ai-runtime/helm/taugrid \
  --version 0.4.2 \
  --namespace tau-system \
  --create-namespace

代码 · 5 行

helm install taugrid \
  oci://mcr.microsoft.com/aks/ai-runtime/helm/taugrid \
  --version 0.4.2 \
  --namespace tau-system \
  --create-namespace

First-party images ship under mcr.microsoft.com/aks/ai-runtime/ for Tau, the TauGrid Portal, and the tau core controller. Microsoft advises pinning versioned tags or immutable digests rather than latest. The CLI installs from GitHub Releases on Linux and macOS, with a PowerShell installer for Windows amd64; the installer verifies the release checksum and does not modify PATH.

官方镜像发布在 mcr.microsoft.com/aks/ai-runtime/ 下，包含 Tau、TauGrid Portal 和 tau 核心控制器。微软建议固定使用带版本号的标签或不可变的摘要（digest），而不是使用 latest。CLI 可从 GitHub Releases 在 Linux 和 macOS 上安装，Windows amd64 提供 PowerShell 安装程序；安装程序会验证发布文件的校验和，且不会修改 PATH 环境变量。

Two operational details matter for anyone evaluating this outside Azure. First, TauGrid sends no telemetry to Microsoft by default, and remote export stays off unless an operator configures a destination. Second, some integrations are still Azure-specific, notably observability through Azure Data Explorer. The stated intent is to support cloud and on-premises Kubernetes without an Azure dependency, and contributions toward that are open.

对于任何在 Azure 之外评估此项目的人来说，有两个操作细节至关重要。首先，TauGrid 默认不向 Microsoft 发送遥测数据，除非操作员配置了目标位置，否则远程导出处于关闭状态。其次，某些集成仍然是 Azure 特有的，特别是通过 Azure Data Explorer 进行的可观测性支持。官方意图是支持与 Azure 无关的云端和本地 Kubernetes，并且为此做出的贡献是开放的。

Key Takeaways

关键要点

Microsoft open-sourced TauGrid on August 28, 2026, under the MIT license at Azure/taugrid.
One Helm install bundles the tau CLI, Kueue queueing, KubeRay orchestration, GPU health monitoring, and observability.
Deployable now on any Kubernetes 1.30+ cluster with GPU nodes, kubectl, and Helm 3.0+.
Evidence records capture config, logs, metrics, and checkpoints, so runs stay reproducible and auditable.
No telemetry by default, but Azure Data Explorer observability remains Azure-specific for now.

Microsoft 于 2026 年 8 月 28 日在 MIT 许可下开源了 TauGrid，代码位于 Azure/taugrid。
一次 Helm 安装即可打包 tau CLI、Kueue 队列管理、KubeRay 编排、GPU 健康监控和可观测性功能。
现在即可部署在任何拥有 GPU 节点、kubectl 和 Helm 3.0+ 的 Kubernetes 1.30+ 集群上。
证据记录捕获配置、日志、指标和检查点，确保运行过程可复现且可审计。
默认情况下无遥测数据，但 Azure Data Explorer 的可观测性目前仍为 Azure 特有。

Check out the AKS Engineering Blog and Azure/taugrid on GitHub. All credit goes to the researcher of this project. Also, feel free to follow us on Twitter and don’t forget to join our 150k+ML SubReddit and Subscribe to our Newsletter. Wait! are you on telegram? now you can join us on telegram as well.

请访问 AKS Engineering Blog 以及 GitHub 上的 Azure/taugrid 仓库。所有功劳归于该项目的研究人员。也欢迎在 Twitter 上关注我们，别忘了加入我们有超过 15 万成员的 ML SubReddit 并订阅我们的新闻通讯。等等！你在 Telegram 上吗？现在你也可以加入我们的 Telegram 群组。

Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.? Connect with us

需要与我们合作推广你的 GitHub 仓库、Hugging Face 页面、产品发布或网络研讨会等吗？请联系我们

The post Microsoft Open-Sources TauGrid: A Kubernetes-Native Stack for GPU AI Workloads appeared first on MarkTechPost.

本文《Microsoft Open-Sources TauGrid: A Kubernetes-Native Stack for GPU AI Workloads》首发于 MarkTechPost。

更进一步：量化金融体系

看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力

进入量化体系 →

话题#Microsoft #TauGrid #Kubernetes #开源