From a5f8bf6d92d14817f7ce5dfeae9832698017840a Mon Sep 17 00:00:00 2001 From: Yuan Fang Date: Tue, 16 Jun 2026 13:31:02 +0800 Subject: [PATCH] docs: add NPU Operator entry --- .../device_management/npu_operator.mdx | 7 +++++++ 1 file changed, 7 insertions(+) create mode 100644 docs/en/infrastructure_management/device_management/npu_operator.mdx diff --git a/docs/en/infrastructure_management/device_management/npu_operator.mdx b/docs/en/infrastructure_management/device_management/npu_operator.mdx new file mode 100644 index 0000000..b7920d5 --- /dev/null +++ b/docs/en/infrastructure_management/device_management/npu_operator.mdx @@ -0,0 +1,7 @@ +# About Alauda Build of NPU Operator + +The Alauda Build of NPU Operator is a Kubernetes operator that automatically manages all software components needed to expose Ascend NPUs to Kubernetes for AI training and inference workloads. It reconciles the full software stack — Ascend driver and firmware, container runtime configuration, MindCluster device plugin, and optional NPU exporter — from a single `NPUOperatorCtl` custom resource. + +With the NPU Operator installed, AI workloads can request `huawei.com/Ascend910` or `huawei.com/Ascend310P` resources the same way they would any other Kubernetes resource. + +