Platform groups operating AI on Kubernetes not often run one factor. They run a queueing system, a distributed runtime, GPU node well being checks, dashboards, and a layer of submission scripts holding all of it collectively. The Azure Kubernetes Service engineering workforce open-sourced TauGrid, which collapses that meeting job right into a single Helm set up.
Is it deployable? Sure, TauGrid is MIT licensed, with container photographs and Helm charts revealed as public OCI artifacts on Microsoft Container Registry. Conditions are a Kubernetes 1.30+ cluster with GPU nodes, kubectl, and Helm 3.0 or later.
What’s TauGrid
TauGrid is a self-hosted platform for operating AI workloads on Kubernetes. It combines 5 issues that platform groups normally combine by hand: the tau CLI, workload queueing and admission by means of Kueue, Ray cluster orchestration by means of KubeRay, node-level GPU well being monitoring, and cluster and workload observability.
The cut up of accountability is the design level. Platform groups personal workspaces, queues, compute profiles, storage, identification, and observability. Researchers work from a repository and the CLI, and submit workloads with out configuring Kubernetes immediately. The codebase is written primarily in Go.
How a job strikes by means of it
A workload is described in a tau.yaml file. The GPU coaching instance revealed by Microsoft runs a PyTorch job on a single A100:
schema_version: 1
identify: aks-gpu-quickstart
run:
entrypoint: practice.py
workload_kind: rayjob
compute:
gpus: 1
staff: 1
cpus: 16
reminiscence: 64Gi
runtime:
picture: mcr.microsoft.com/aks/ai-runtime/ray:py3.12-ray2.56.0-cuda13.0
pip:
- torch>=2.4.0On tau run, TauGrid resolves platform coverage, renders a Kubernetes Job or a KubeRay RayJob, and submits it by means of Kueue. The six levels Microsoft paperwork are submission, queueing, execution, monitoring, restoration, and proof. Restoration covers retry, resume from checkpoint, and failure analysis. Proof information seize workload metadata, configuration, logs, metrics, checkpoints, and execution historical past, which is what makes a run reproducible and auditable later.
When a number of groups share a cluster, their jobs land in a shared Kueue ClusterQueue. Kueue admits every one on quota and precedence, and Kubernetes locations it on wholesome GPUs.
Interactive explainer
Set up is a Helm chart pulled straight from MCR:
helm set up taugrid
oci://mcr.microsoft.com/aks/ai-runtime/helm/taugrid
--version 0.4.2
--namespace tau-system
--create-namespaceFirst-party photographs ship underneath mcr.microsoft.com/aks/ai-runtime/ for Tau, the TauGrid Portal, and the tau core controller. Microsoft advises pinning versioned tags or immutable digests relatively than newest. The CLI installs from GitHub Releases on Linux and macOS, with a PowerShell installer for Home windows amd64; the installer verifies the discharge checksum and doesn’t modify PATH.
Two operational particulars matter for anybody evaluating this exterior Azure. First, TauGrid sends no telemetry to Microsoft by default, and distant export stays off except an operator configures a vacation spot. Second, some integrations are nonetheless Azure-specific, notably observability by means of Azure Knowledge Explorer. The acknowledged intent is to assist cloud and on-premises Kubernetes with out an Azure dependency, and contributions towards which are open.
Key Takeaways
- Microsoft open-sourced TauGrid on August 28, 2026, underneath the MIT license at Azure/taugrid.
- One Helm set up bundles the
tauCLI, Kueue queueing, KubeRay orchestration, GPU well being monitoring, and observability. - Deployable now on any Kubernetes 1.30+ cluster with GPU nodes, kubectl, and Helm 3.0+.
- Proof information seize config, logs, metrics, and checkpoints, so runs keep reproducible and auditable.
- No telemetry by default, however Azure Knowledge Explorer observability stays Azure-specific for now.
Try the AKS Engineering Weblog and Azure/taugrid on GitHub. All credit score goes to the researcher of this challenge. Additionally, be at liberty to comply with us on Twitter and don’t overlook to affix our 150k+ML SubReddit and Subscribe to our E-newsletter. Wait! are you on telegram? now you possibly can be a part of us on telegram as nicely.
Have to companion with us for selling your GitHub Repo OR Hugging Face Web page OR Product Launch OR Webinar and so forth.? Join with us
Asif Razzaq is the CEO of Marktechpost AI Media Inc.. As a visionary entrepreneur and engineer, Asif is dedicated to harnessing the potential of Synthetic Intelligence for social good. His most up-to-date endeavor is the launch of an Synthetic Intelligence Media Platform, Marktechpost, which stands out for its in-depth protection of machine studying and deep studying information that’s each technically sound and simply comprehensible by a large viewers. The platform boasts of over 2 million month-to-month views, illustrating its reputation amongst audiences.
