diff --git a/.github/workflows/fern-docs-ci.yml b/.github/workflows/fern-docs-ci.yml index de629d340..7ed973be9 100644 --- a/.github/workflows/fern-docs-ci.yml +++ b/.github/workflows/fern-docs-ci.yml @@ -63,9 +63,3 @@ jobs: - name: Fern check run: fern check - - - name: Check links - uses: lycheeverse/lychee-action@8646ba30535128ac92d33dfc9133794bfdd9b411 # v2.8.0 - with: - args: --offline --no-progress 'docs/**/*.md' - fail: true diff --git a/docs/user/cluster-management/index.md b/docs/user/cluster-management/index.md index 815c30da7..68bd405ae 100644 --- a/docs/user/cluster-management/index.md +++ b/docs/user/cluster-management/index.md @@ -29,7 +29,7 @@ After installing NVCA on a cluster: - If your cloud provider does not support the NVIDIA GPU Operator, [Manual Instance Configuration](./configuration.md) is possible, but not recommended due to lack of maintainability. - To get the most out of clusters with multi-node NVLink ([MNNVL](https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/dra-cds.html#dra-docs-compute-domains)) GPUs like [GB200](https://www.nvidia.com/en-us/data-center/gb200-nvl72/), the [NVIDIA GPU DRA driver](https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/dra-intro-install.html) must be installed. See the [nvlink-optimized-clusters](./configuration.md) for details. - - For development or testing environments without physical GPUs, install the [fake-gpu-operator](../fake-gpu-operator.md) instead. + - For development or testing environments without physical GPUs, install the [fake-gpu-operator](../fake-gpu-operator) instead. - Registering the cluster requires `kubectl` and `helm` installed. diff --git a/docs/user/cluster-management/self-managed.md b/docs/user/cluster-management/self-managed.md index bb0030796..7d3ae5a57 100644 --- a/docs/user/cluster-management/self-managed.md +++ b/docs/user/cluster-management/self-managed.md @@ -29,7 +29,7 @@ Before installing the NVCA Operator, ensure the following prerequisites are met: - The [control plane](../helmfile-installation.md) is installed and all core services are running. -- The [NVIDIA GPU Operator](https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/getting-started.html) is installed on the GPU cluster. The GPU Operator manages the NVIDIA drivers, device plugin, and GPU feature discovery required for workload scheduling. For development or testing environments without physical GPUs, see [fake-gpu-operator](../fake-gpu-operator.md). +- The [NVIDIA GPU Operator](https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/getting-started.html) is installed on the GPU cluster. The GPU Operator manages the NVIDIA drivers, device plugin, and GPU feature discovery required for workload scheduling. For development or testing environments without physical GPUs, see [fake-gpu-operator](../fake-gpu-operator). - (Optional) The [KAI Scheduler](./kai-scheduler.md) can be installed on the GPU cluster for optimized AI workload scheduling and bin-packing of GPU resources. It is required only when you enable the `KAIScheduler` feature flag. See [NVCA Configuration](./configuration.md). diff --git a/docs/user/infrastructure-sizing.md b/docs/user/infrastructure-sizing.md index eca83d8a2..1d4c83aa1 100644 --- a/docs/user/infrastructure-sizing.md +++ b/docs/user/infrastructure-sizing.md @@ -7,8 +7,8 @@ types and storage classes are AWS examples. For another cloud service provider storage classes that provide at least the listed vCPU, memory, storage, zone spread, and GPU compatibility characteristics. Actual requirements depend on workload characteristics, function count, and request concurrency. Use the -[self-managed-grpc-load-test](./grpc-load-testing.md) and -[self-managed-http-load-test](./http-load-testing.md) guides to validate +[self-managed-grpc-load-test](../g-rpc-load-testing) and +[self-managed-http-load-test](../http-load-testing) guides to validate throughput and tune your control plane accordingly. @@ -56,7 +56,7 @@ For local development of the stack or functions, CI pipelines, or quick demos, you can run the entire NVCF stack on a single machine using k3d. This setup uses a single Cassandra replica, fake GPUs, and ephemeral `local-path` storage. -See [local-development](./local-development.md) for full step-by-step instructions. +See [local-development](../local-development) for full step-by-step instructions. ### Staging / Demo @@ -80,7 +80,7 @@ Use this tier for: You can also run the full stack on your laptop using Kind or k3d. See -[local-development](./local-development.md) for instructions. +[local-development](../local-development) for instructions. @@ -153,7 +153,7 @@ GPU requirements: - Physical GPU hardware on worker nodes For development and testing environments without GPUs, install the fake GPU -operator to simulate GPU resources. See [fake-gpu-operator](./fake-gpu-operator.md) for +operator to simulate GPU resources. See [fake-gpu-operator](../fake-gpu-operator) for instructions. ## Storage Recommendations @@ -180,8 +180,8 @@ The default control-plane resource sizing shipped with the helmfile stack is designed to handle approximately 100 concurrent users. If you need higher throughput: -1. Benchmark your deployment using the [self-managed-grpc-load-test](./grpc-load-testing.md) - or [self-managed-http-load-test](./http-load-testing.md) guide. Start with +1. Benchmark your deployment using the [self-managed-grpc-load-test](../g-rpc-load-testing) + or [self-managed-http-load-test](../http-load-testing) guide. Start with `--vus 100` and increase gradually. 2. Scale node pools independently. Cassandra, OpenBao, and control-plane pools can each be scaled without affecting the others. @@ -191,7 +191,7 @@ throughput: - [Quickstart](./quickstart.md): One-click fresh installation walkthrough -- [self-managed-grpc-load-test](./grpc-load-testing.md): Validate control-plane throughput -- [self-managed-http-load-test](./http-load-testing.md): Validate HTTP invocation throughput +- [self-managed-grpc-load-test](../g-rpc-load-testing): Validate control-plane throughput +- [self-managed-http-load-test](../http-load-testing): Validate HTTP invocation throughput diff --git a/docs/user/installation.md b/docs/user/installation.md index c5e56b94d..c6a833ed0 100644 --- a/docs/user/installation.md +++ b/docs/user/installation.md @@ -9,7 +9,7 @@ For a full list of required artifacts, see [self-hosted-artifact-manifest](./man ![Self-hosted component overview](images/nvcf-high-level-stack.svg) -Want to try NVCF locally first? See [Local Development](./local-development.md) to create a k3d cluster, then use the [Quickstart](./quickstart.md) local k3d flow. +Want to try NVCF locally first? See [Local Development](../local-development) to create a k3d cluster, then use the [Quickstart](./quickstart.md) local k3d flow. @@ -75,7 +75,7 @@ See [NVIDIA GPU Operator documentation](https://docs.nvidia.com/datacenter/cloud Fake GPU Operator for development and testing: For environments without actual GPU hardware, install the fake GPU operator to simulate -GPU resources. See [fake-gpu-operator](./fake-gpu-operator.md) for full instructions. +GPU resources. See [fake-gpu-operator](../fake-gpu-operator) for full instructions. #### SMB CSI Driver diff --git a/docs/user/quickstart.md b/docs/user/quickstart.md index 42e49ba6b..451c8db7e 100644 --- a/docs/user/quickstart.md +++ b/docs/user/quickstart.md @@ -346,7 +346,7 @@ Common local k3d issues: ## See Also -- [Local Development](./local-development.md) for local k3d variants and cleanup commands. +- [Local Development](../local-development) for local k3d variants and cleanup commands. - [Helmfile Installation](./helmfile-installation.md) for remote or manual control-plane installs. - [Self-Managed Clusters](./cluster-management/self-managed.md) for registering GPU clusters outside the local quickstart. - `src/clis/nvcf-cli/examples/` in this repository for sample CLI input files. diff --git a/fern/docs.yml b/fern/docs.yml index 7eeed3fb4..3cdb5a79b 100644 --- a/fern/docs.yml +++ b/fern/docs.yml @@ -36,6 +36,10 @@ versions: path: versions/v0.5.yml slug: "v0.5" +check: + rules: + broken-links: error + redirects: - source: "/nvcf/index.html" destination: "/nvcf/"