ssiog/launch/job-set.yaml (133 lines of code) (raw):

# Copyright 2024 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: jobset.x-k8s.io/v1alpha2 kind: JobSet metadata: generateName: ssiog-training- annotations: alpha.jobset.sigs.k8s.io/exclusive-topology: cloud.google.com/gke-nodepool # 1:1 job replica to node pool assignment spec: failurePolicy: maxRestarts: 0 replicatedJobs: - name: main replicas: 1 template: spec: parallelism: 2 # Should be smaller than the number of VMs completions: 2 # Same as the above. backoffLimit: 0 # When any pod fails, the job is failed template: metadata: # Required for GCSFuse. # For other storage solutions, please modify this section. annotations: gke-gcsfuse/volumes: "true" gke-gcsfuse/cpu-limit: "0" gke-gcsfuse/memory-limit: "0" gke-gcsfuse/ephemeral-storage-limit: "0" spec: # The main entry point. Use the same image as the benchmark. containers: - name: ssiog-benchmark image: us-west1-docker.pkg.dev/gcs-tess/ssiog-training/v0.8.0@sha256:48e36ae4f0920d91951b8fef39e55ddc796a11ed7acfc9b88efd2e8555b9b718 env: - name: REPLICATED_JOB_NAME valueFrom: fieldRef: fieldPath: metadata.annotations['jobset.sigs.k8s.io/replicatedjob-name'] - name: JOB_INDEX valueFrom: fieldRef: fieldPath: metadata.annotations['jobset.sigs.k8s.io/job-index'] - name: JOBSET_NAME valueFrom: fieldRef: fieldPath: metadata.annotations['jobset.sigs.k8s.io/jobset-name'] - name: JOB_COMPLETION_INDEX valueFrom: fieldRef: fieldPath: metadata.annotations['batch.kubernetes.io/job-completion-index'] ports: - containerPort: 5670 securityContext: privileged: true command: - /bin/bash - -c - | mkdir -p /mnt/benchmark-output/test_0-8-0-0/${JOBSET_NAME} || true mkdir -p /output/ || true args=( --prefix /mnt/benchmark-inputs --object-count-limit=16 --epochs=2 --background-threads=8 --sample-size=65536 --steps=50 --batch-size=32 --group-member-id=${JOB_COMPLETION_INDEX} --group-coordinator-address=${JOBSET_NAME}-${REPLICATED_JOB_NAME}-0-0.${JOBSET_NAME} --group-coordinator-port=5670 --group-size=2 --log-metrics=True --metrics-file=/output/results-${JOB_COMPLETION_INDEX}.csv --log-level=INFO --read-order=FullRandom --export-metrics=True --exporter-type=cloud --label=test_0-8-0-0 ) /app/training.py "${args[@]}" echo "Copying the local metrics to bucket..." cp -r /output/* /mnt/benchmark-output/test_0-8-0-0/${JOBSET_NAME}/ volumeMounts: - mountPath: /mnt/benchmark-output name: gcsfuse-outputs readOnly: false - mountPath: /mnt/benchmark-inputs name: mnt-inputs readOnly: true initContainers: - name: capture-system-info image: ubuntu:24.04 restartPolicy: Always command: - /bin/bash - -c - | echo "nproc"; nproc; echo; echo echo "Memory"; free -h; echo; echo echo "Mounted"; df; echo; echo function sigterm() { exit 0; } trap sigterm SIGTERM tail -f /dev/null volumeMounts: # For other storage solutions, please modify the mount path and specify it in the # `dataset_directory` argument in the command above. - mountPath: /mnt/benchmark-output name: gcsfuse-outputs readOnly: false schedulerName: default-scheduler restartPolicy: Never affinity: nodeAffinity: requiredDuringSchedulingIgnoredDuringExecution: nodeSelectorTerms: - matchExpressions: - key: cloud.google.com/gke-nodepool operator: NotIn values: - default-pool hostNetwork: true dnsPolicy: ClusterFirstWithHostNet terminationGracePeriodSeconds: 30 # Set up k8s SA to access the metrics bucket: https://cloud.google.com/kubernetes-engine/docs/how-to/persistent-volumes/cloud-storage-fuse-csi-driver#authentication. # For GCSFuse, also set up access to the dataset bucket. serviceAccountName: ssiog-runner-ksa volumes: - name: gcsfuse-outputs csi: driver: gcsfuse.csi.storage.gke.io volumeAttributes: bucketName: princer-ssiog-metrics-bkt - name: mnt-inputs csi: driver: gcsfuse.csi.storage.gke.io readOnly: false volumeAttributes: bucketName: princer-ssiog-data-bkt mountOptions: "debug_fuse,implicit-dirs,metadata-cache:ttl-secs:-1,metadata-cache:stat-cache-max-size-mb:-1,metadata-cache:type-cache-max-size-mb:-1"