Add llama3-1-70b-2node-bf16-seq8192-gbs2048 recipes on A4

training/a4/llama3-1-70b/nemo-pretraining-gke/2node-bf16-seq8192-gbs2048/recipe/Chart.yaml

@@ -0,0 +1,20 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+apiVersion: v2
+name: a4_jobset_workload
+description: a4_jobset_workload
+type: application
+version: 0.1.0
+appVersion: "1.16.0"

training/a4/llama3-1-70b/nemo-pretraining-gke/2node-bf16-seq8192-gbs2048/recipe/README.md

@@ -0,0 +1,153 @@
+<!-- mdformat global-off -->
+# Pretrain llama3-1-70b-seq8192-gbs2048-mbs1-gpus16 workloads on a4 GKE Node pools with Nvidia NeMo Framework
+
+This recipe outlines the steps for running a llama3-1-70b-seq8192-gbs2048-mbs1-gpus16 pretraining
+workload on [a4 GKE Node pools](https://cloud.google.com/kubernetes-engine) by using the
+[NVIDIA NeMo framework](https://github.com/NVIDIA/nemo).
+
+## Orchestration and deployment tools
+
+For this recipe, the following setup is used:
+
+- Orchestration - [Google Kubernetes Engine (GKE)](https://cloud.google.com/kubernetes-engine)
+- Pretraining job configuration and deployment - A Helm chart is used to
+  configure and deploy the [Kubernetes Jobset](https://kubernetes.io/blog/2025/03/23/introducing-jobset) resource which manages the execution of the
+  [NeMo pretraining workload](https://github.com/NVIDIA/nemo).
+
+## Test environment
+
+This recipe has been optimized for and tested with the following configuration:
+
+- GKE cluster
+Please follow Cluster Toolkit [instructions](https://github.com/GoogleCloudPlatform/cluster-toolkit/tree/main/examples/gke-a4)
+to create your a4 GKE cluster.
+
+## Training dataset
+
+This recipe uses a mock pretraining dataset provided by the NeMo framework.
+
+## Docker container image
+
+This recipe uses the following docker images:
+
+- `nvcr.io/nvidia/nemo:25.07`
+- `us-docker.pkg.dev/gce-ai-infra/gpudirect-gib/nccl-plugin-gib:v1.1.0`
+
+## Run the recipe
+
+From your client workstation, complete the following steps:
+
+### Configure environment settings
+
+Set the environment variables to match your environment:
+
+ ```bash
+ export PROJECT_ID=<PROJECT_ID>
+ export CLUSTER_REGION=<CLUSTER_REGION>
+ export CLUSTER_NAME=<CLUSTER_NAME>
+ export GCS_BUCKET=<GCS_BUCKET> # Note: path should not be prefixed with gs://
+ export KUEUE_NAME=<KUEUE_NAME>
+ ```
+
+Replace the following values:
+
+ - `<PROJECT_ID>`: your Google Cloud project ID.
+ - `<CLUSTER_REGION>`: the region where your cluster is located.
+ - `<CLUSTER_NAME>`: the name of your GKE cluster.
+ - `<GCS_BUCKET>`: the name of your Cloud Storage bucket. Don't include the `gs://` prefix.
+ - `<KUEUE_NAME>`: the name of the Kueue local queue. The default queue created by the cluster toolkit is `a4`. Make sure to verify the name of the local queue in your cluster.
+
+Set the default project:
+
+ ```bash
+ gcloud config set project $PROJECT_ID
+ ```
+
+### Get the recipe
+
+Clone the `gpu-recipes` repository and set a reference to the recipe folder.
+
+```
+git clone https://github.com/ai-hypercomputer/gpu-recipes.git
+cd gpu-recipes
+export REPO_ROOT=`git rev-parse --show-toplevel`
+export RECIPE_ROOT=$REPO_ROOT/training/a4/llama3-1-70b-seq8192-gbs2048-mbs1-gpus16/nemo-pretraining-gke/2_nodes
+cd $RECIPE_ROOT
+```
+
+### Get cluster credentials
+
+```
+gcloud container clusters get-credentials $CLUSTER_NAME --region $CLUSTER_REGION
+```
+
+### Configure and submit a pretraining job
+
+#### Using 2 node (16 gpus) bf16 precision
+To execute the job with the default settings, run the following command from
+your client:
+
+    ```bash
+    cd $RECIPE_ROOT
+    export WORKLOAD_NAME=$USER-a4-llama3-1-70b-seq8192-gbs2048-mbs1-gpus16-2node
+    helm install $WORKLOAD_NAME . -f values.yaml \
+    --set-file workload_launcher=launcher.sh \
+    --set-file workload_config=llama3-1-70b-seq8192-gbs2048-mbs1-gpus16.py \
+    --set workload.image=nvcr.io/nvidia/nemo:25.07 \
+    --set volumes.gcsMounts[0].bucketName=${GCS_BUCKET} \
+    --set volumes.gcsMounts[0].mountPath=/job-logs \
+    --set workload.envs[0].value=/job-logs/$WORKLOAD_NAME \
+    --set queue=${KUEUE_NAME}
+    ```
+
+**Examples**
+
+-   To set the number of training steps to 100, run the following command from
+    your client:
+
+    ```bash
+    cd $RECIPE_ROOT
+    export WORKLOAD_NAME=$USER-a4-llama3-1-70b-seq8192-gbs2048-mbs1-gpus16-2node
+    helm install $WORKLOAD_NAME . -f values.yaml \
+    --set-file workload_launcher=launcher.sh \
+    --set-file workload_config=llama3-1-70b-seq8192-gbs2048-mbs1-gpus16.py \
+    --set workload.image=nvcr.io/nvidia/nemo:25.07 \
+    --set volumes.gcsMounts[0].bucketName=${GCS_BUCKET} \
+    --set volumes.gcsMounts[0].mountPath=/job-logs \
+    --set workload.envs[0].value=/job-logs/$WORKLOAD_NAME \
+    --set queue=${KUEUE_NAME} \
+    --set workload.arguments[0]="trainer.max_steps=100"
+    ```
+
+### Monitor the job
+
+To check the status of pods in your job, run the following command:
+
+```
+kubectl get pods | grep $USER-a4-llama3-1-70b-seq8192-gbs2048-mbs1-gpus16-2node
+```
+
+Replace the following:
+
+- JOB_NAME_PREFIX - your job name prefix. For example $USER-a4-llama3-1-70b-seq8192-gbs2048-mbs1-gpus16-2node.
+
+To get the logs for one of the pods, run the following command:
+
+```
+kubectl logs POD_NAME
+```
+
+Information about the training job's progress, including crucial details such as
+loss, step count, and step time, is generated by the rank 0 process.
+This process runs on the pod whose name begins with
+`JOB_NAME_PREFIX-workload-0-0`.
+For example: `$USER-a4-llama3-1-70b-seq8192-gbs2048-mbs1-gpus16-2node-workload-0-0-s9zrv`.
+
+### Uninstall the Helm release
+
+You can delete the job and other resources created by the Helm chart. To
+uninstall Helm, run the following command from your client:
+
+```bash
+helm uninstall $USER-a4-llama3-1-70b-seq8192-gbs2048-mbs1-gpus16-2node
+```

training/a4/llama3-1-70b/nemo-pretraining-gke/2node-bf16-seq8192-gbs2048/recipe/launcher.sh

@@ -0,0 +1,105 @@
+usage()
+{
+cat << EOF
+usage: bash ./launcher.sh [config-override  [config-override ...]]
+config-override  (Optional) A  NeMo configuration override. E.g. trainer.max_steps=10000.
+EOF
+}
+
+parse_args() {
+  while [ "$1" != "" ]; do
+    case $(grep -o "=" <<< "$1" | wc -l) in
+        1  )
+        config_overrides+=("$1")
+        ;;
+        * )
+            echo "Invalid config override: $1"
+            usage
+            exit 1
+    esac
+    shift
+  done
+  config_overrides="${config_overrides[*]}"
+}
+
+config_overrides=()
+parse_args "$@"
+
+if [ -z "${config_overrides}" ]; then
+  echo "No NeMo config overrides specified"
+else
+  echo "NeMo config overrides:"
+  echo "  ${config_overrides}"
+fi
+
+export LD_LIBRARY_PATH="$NCCL_PLUGIN_PATH"
+ldconfig $LD_LIBRARY_PATH
+echo "Added $LD_LIBRARY_PATH to ldconfig:"
+ldconfig -p | grep libcuda | sed 's/^/  /'
+echo ""
+
+if [[ -n "${EXPLICIT_LOG_DIR}" ]]; then
+  explicit_log_dir=${EXPLICIT_LOG_DIR}
+else
+  explicit_log_dir=workload_logs
+fi
+echo "Logging to ${explicit_log_dir}"
+
+if [[ -n "${TOKENIZER_PATH}" ]]; then
+  echo "Getting tokenizer files"
+  cp ${TOKENIZER_PATH}/* .
+  echo ""
+fi
+
+echo "Launching Torch distributed on the node rank $JOB_COMPLETION_INDEX out of $NNODES nodes"
+
+
+# Update nemo run so we can export the config.
+pip install git+https://github.com/NVIDIA/NeMo-Run.git@6550ff68204e5095452098eed3765ed765de5d33
+pip install git+https://github.com/NVIDIA/dllogger#egg=dllogger
+
+
+# Export the nemo2 config to yaml.
+python ${NEMO_LAUNCH_SCRIPT} --factory "recipe()" \
+trainer.num_nodes="$NNODES" \
+log.explicit_log_dir="${explicit_log_dir}" \
+trainer.max_steps=25 \
+trainer.num_nodes=2 \
+trainer.devices=8 \
+${config_overrides} \
+--to-yaml exported_nemo_config.yaml
+
+# Create the nsys directory.
+mkdir -p ${explicit_log_dir}/nsys
+
+OMP_NUM_THREADS=12 NSYS_CONFIG_DIRECTIVES="AgentLaunchTimeoutSec=240;AppLaunchTimeoutSec=240" TORCH_NCCL_ENABLE_MONITORING=0 \
+/usr/local/bin/nsys profile -s none -t nvtx,cuda --capture-range=cudaProfilerApi --capture-range-end=stop \
+-o ${explicit_log_dir}/nsys/noderank-${JOB_COMPLETION_INDEX} \
+--session-new "nemo-rank${JOB_COMPLETION_INDEX}"-$RANDOM \
+--wait all \
+torchrun \
+--nproc-per-node="${GPUS_PER_NODE}" \
+--nnodes="${NNODES}" \
+--node_rank="${JOB_COMPLETION_INDEX}" \
+--rdzv_id="${JOB_IDENTIFIER}" \
+--master_addr="${MASTER_ADDR}" \
+--master_port="${MASTER_PORT}" \
+${NEMO_LAUNCH_SCRIPT} --factory "recipe()" \
+trainer.num_nodes="$NNODES" \
+log.explicit_log_dir="${explicit_log_dir}" \
+trainer.max_steps=25 \
+trainer.num_nodes=2 \
+trainer.devices=8 \
+${config_overrides}
+
+if [[ "$JOB_COMPLETION_INDEX" == "0" ]]; then
+  mkdir -p ${ARTIFACT_DIR}
+  cp -r ${explicit_log_dir}/* ${ARTIFACT_DIR}/
+  cp ${NEMO_LAUNCH_SCRIPT} ${ARTIFACT_DIR}/run-cli.py
+  cp dllogger.json ${ARTIFACT_DIR}/dllogger.json
+  cp exported_nemo_config.yaml ${ARTIFACT_DIR}/nemo-configuration.yaml
+  env > ${ARTIFACT_DIR}/environ.txt
+  ls ${ARTIFACT_DIR}
+fi
+echo "Training completed"
+echo "Pod on $(hostname --fqdn) is exiting"