‫Google משתמשת בטכנולוגיית AI כדי לתרגם תוכן לשפה המועדפת עליך. בתרגומים כאלו עשויות להיות שגיאות.

הפעלת NCCL באשכולות GKE בהתאמה אישית עם A3 Mega או A3 High

בדף הזה מוסבר איך להריץ בדיקות של NVIDIA Collective Communications Library (NCCL) באשכולות GKE בהתאמה אישית שמשתמשים בפרוטוקולי הרשת GPUDirect-TCPXO ו-GPUDirect-TCPX. אשכול GKE בהתאמה אישית הוא אשכול שיוצרים באמצעות פקודות gcloud.

אפשר להשתמש בבדיקות שמתוארות בדף הזה בתרחישים הבאים:

אם באשכול GKE שלכם נעשה שימוש בצמתים עם הפעלה גמישה, צריך להשתמש בבדיקה בסיסית בשני צמתים.
אם באשכול GKE שלכם נעשה שימוש בסוגים שונים של צמתים, כמו צמתים לפי דרישה או צמתים שמוגבלים להזמנה, צריך להשתמש בבדיקת NCCL עם תזמון שמודע לטופולוגיה.

לפני שמתחילים

הבדיקות בדף הזה משתמשות ב-JobSet וב-Kueue עם Topology Aware Scheduling (TAS). לפני שמריצים בדיקות, צריך להגדיר את האשכול ולבצע את הפעולות הבאות:

מתקינים את JobSet.

מתקינים את Kueue.

kubectl apply --server-side -f https://github.com/kubernetes-sigs/kueue/releases/download/v0.16.5/manifests.yaml

הגדרת האשכול באמצעות Jobset ו-Kueue

אחרי שמתקינים את JobSet ואת Kueue, מבצעים את השלבים הבאים:

שומרים את קובץ המניפסט הבא בשם kueue-config.yaml:

A3 High

apiVersion: kueue.x-k8s.io/v1beta2
kind: Topology
metadata:
  name: "gke-default"
spec:
  levels:
  - nodeLabel: "cloud.google.com/gce-topology-block"
  - nodeLabel: "cloud.google.com/gce-topology-subblock"
  - nodeLabel: "cloud.google.com/gce-topology-host"
  - nodeLabel: "kubernetes.io/hostname"
---
apiVersion: kueue.x-k8s.io/v1beta2
kind: ResourceFlavor
metadata:
  name: a3-high-flavor
spec:
  nodeLabels:
    cloud.google.com/gke-accelerator: nvidia-h100-80gb
  topologyName: "gke-default"
---
apiVersion: kueue.x-k8s.io/v1beta2
kind: ResourceFlavor
metadata:
  name: a3-high-dws-flavor
spec:
  nodeLabels:
    cloud.google.com/gke-accelerator: nvidia-h100-80gb
  topologyName: "gke-default"
  tolerations:
  - key: "cloud.google.com/gke-queued"
    operator: "Exists"
    effect: NoSchedule
---     
apiVersion: kueue.x-k8s.io/v1beta2
kind: AdmissionCheck
metadata:
  name: dws-prov
spec:
  controllerName: kueue.x-k8s.io/provisioning-request
  parameters:
    apiGroup: kueue.x-k8s.io
    kind: ProvisioningRequestConfig
    name: dws-config
---
apiVersion: kueue.x-k8s.io/v1beta2
kind: ProvisioningRequestConfig
metadata:
  name: dws-config
spec:
  provisioningClassName: queued-provisioning.gke.io
  podSetUpdates:
  - key: autoscaling.gke.io/provisioning-request
    valueFromProvisioningClassDetail: ResizeRequestName
  managedResources:
  - nvidia.com/gpu
---
apiVersion: kueue.x-k8s.io/v1beta2
kind: ClusterQueue
metadata:
  name: cq-tas
spec:
  namespaceSelector: {}
  clusterQueueingStrategy: BestEffortFIFO
  resourceGroups:
  - flavors:
    - name: a3-high-flavor
      resources:
      - name: "cpu"
        nominalQuota: 1000
      - name: "memory"
        nominalQuota: 1000Ti
      - name: "nvidia.com/gpu"
        nominalQuota: 1000
    - name: a3-high-dws-flavor
      resources:
      - name: "cpu"
        nominalQuota: 1000
      - name: "memory"
        nominalQuota: 1000Ti
      - name: "nvidia.com/gpu"
        nominalQuota: 1000
  admissionChecksStrategy:
    admissionChecks:
    - name: "dws-prov"
      onFlavors: [a3-high-dws-flavor]
---
apiVersion: kueue.x-k8s.io/v1beta2
kind: LocalQueue
metadata:
  namespace: default
  name: lq-tas
spec:
  clusterQueue: cq-tas

A3 Mega

apiVersion: kueue.x-k8s.io/v1beta2
kind: Topology
metadata:
  name: "gke-default"
spec:
  levels:
  - nodeLabel: "cloud.google.com/gce-topology-block"
  - nodeLabel: "cloud.google.com/gce-topology-subblock"
  - nodeLabel: "cloud.google.com/gce-topology-host"
  - nodeLabel: "kubernetes.io/hostname"
---
apiVersion: kueue.x-k8s.io/v1beta2
kind: ResourceFlavor
metadata:
  name: a3-mega-flavor
spec:
  nodeLabels:
    cloud.google.com/gke-accelerator: nvidia-h100-mega-80gb
  topologyName: "gke-default"
---
apiVersion: kueue.x-k8s.io/v1beta2
kind: ResourceFlavor
metadata:
  name: a3-mega-dws-flavor
spec:
  nodeLabels:
    cloud.google.com/gke-accelerator: nvidia-h100-mega-80gb
  topologyName: "gke-default"
  tolerations:
  - key: "cloud.google.com/gke-queued"
    operator: "Exists"
    effect: NoSchedule
---     
apiVersion: kueue.x-k8s.io/v1beta2
kind: AdmissionCheck
metadata:
  name: dws-prov
spec:
  controllerName: kueue.x-k8s.io/provisioning-request
  parameters:
    apiGroup: kueue.x-k8s.io
    kind: ProvisioningRequestConfig
    name: dws-config
---
apiVersion: kueue.x-k8s.io/v1beta2
kind: ProvisioningRequestConfig
metadata:
  name: dws-config
spec:
  provisioningClassName: queued-provisioning.gke.io
  podSetUpdates:
  - key: autoscaling.gke.io/provisioning-request
    valueFromProvisioningClassDetail: ResizeRequestName
  managedResources:
  - nvidia.com/gpu
---
apiVersion: kueue.x-k8s.io/v1beta2
kind: ClusterQueue
metadata:
  name: cq-tas
spec:
  namespaceSelector: {}
  clusterQueueingStrategy: BestEffortFIFO
  resourceGroups:
  - flavors:
    - name: a3-mega-flavor
      resources:
      - name: "cpu"
        nominalQuota: 1000
      - name: "memory"
        nominalQuota: 1000Ti
      - name: "nvidia.com/gpu"
        nominalQuota: 1000
    - name: a3-mega-dws-flavor
      resources:
      - name: "cpu"
        nominalQuota: 1000
      - name: "memory"
        nominalQuota: 1000Ti
      - name: "nvidia.com/gpu"
        nominalQuota: 1000
  admissionChecksStrategy:
    admissionChecks:
    - name: "dws-prov"
      onFlavors: [a3-mega-dws-flavor]
---
apiVersion: kueue.x-k8s.io/v1beta2
kind: LocalQueue
metadata:
  namespace: default
  name: lq-tas
spec:
  clusterQueue: cq-tas

החלת המניפסט:
```
kubectl apply -f kueue-config.yaml
```

כשמריצים עומסי עבודה עם TAS מופעל, אפשר לציין עד כמה מגבילים אילוצים של טופולוגיה באמצעות אחת מההערות הבאות במניפסט של עומס העבודה:

‫kueue.x-k8s.io/podset-required-topology: אם משתמשים בהערה הזו, Kueue חוסם את התזמון עד שאפשר לתזמן את עומס העבודה במסגרת מגבלת הטופולוגיה המבוקשת. אפשר להשתמש בהערה הזו כדי לוודא שקבוצות של פודים ממוקמות יחד כדי להשיג ביצועים אופטימליים.
‫kueue.x-k8s.io/podset-preferred-topology: אם משתמשים בהערה הזו, Kueue מנסה לתזמן את הפודים בהתאם לאילוץ הטופולוגיה שצוין, אבל אם זה לא אפשרי, הוא מאשר את עומס העבודה בלי לעמוד באילוצי הטופולוגיה.

הערה: לא מומלץ להשתמש במצב הנדרש עם DWS Flex-start. בגלל ש-Flex-start מספקת צמתים באופן דינמי, יכול להיות שהצמתים שיתקבלו לא יעמדו בדרישות טופולוגיות מחמירות, מה שעלול לגרום לכך שלא ניתן יהיה לתזמן עומסי עבודה. במקום זאת, צריך להשתמש ב-podset-preferred-topology.

לכל אחת מההערות, מציינים אחד מהערכים הבאים כמגבלת הטופולוגיה:

‫cloud.google.com/gce-topology-block: מתזמן פודים באותו בלוק רשת.
‫cloud.google.com/gce-topology-subblock: מתזמן פודים באותו מתלה.
‫cloud.google.com/gce-topology-host: מתזמן פודים באותו מארח פיזי.

בדיקה בשני צמתים עם התחלה גמישה

כדי להריץ בדיקות NCCL באשכול GKE שמשתמש במכונות וירטואליות מסוג A3 Mega או A3 High Flex-start, צריך לפעול לפי ההליך הבא. בפרוצדורה הזו נעשה שימוש במניפסט JobSet כדי להריץ בדיקת NCCL בשני צמתים.

שומרים את קובץ המניפסט הבא בשם nccl-tas-jobset.yaml:

A3 Mega

apiVersion: v1
kind: ConfigMap
metadata:
  name: nccl-configmap
data:
  allgather.sh: |
    #!/bin/bash
    service ssh restart;
    /scripts/init_ssh.sh ${@};
    pushd /scripts;
    /scripts/gen_hostfiles.sh ${@};
    popd;
    # Set up environment variables for GPUDirect-TCPXO
    export LD_LIBRARY_PATH=/usr/local/nvidia/lib64
    export NCCL_FASTRAK_CTRL_DEV=eth0
    export NCCL_FASTRAK_IFNAME=eth1,eth2,eth3,eth4,eth5,eth6,eth7,eth8
    export NCCL_SOCKET_IFNAME=eth0
    export NCCL_CROSS_NIC=0
    export NCCL_ALGO=Ring,Tree
    export NCCL_PROTO=Simple
    export NCCL_NET_GDR_LEVEL=PIX
    # Run the benchmark
    /scripts/demo-run-nccl-test-tcpxo-via-mpi.sh
---
apiVersion: jobset.x-k8s.io/v1alpha2
kind: JobSet
metadata:
  name: nccl-tas-test
  labels:
    kueue.x-k8s.io/queue-name: lq-tas
spec:
  ttlSecondsAfterFinished: 1200
  suspend: true
  network:
    enableDNSHostnames: true
  replicatedJobs:
    - name: worker
      replicas: 2
      template:
        spec:
          parallelism: 1
          completions: 1
          template:
            metadata:
              annotations:
                kueue.x-k8s.io/podset-preferred-topology: "cloud.google.com/gce-topology-block"
                networking.gke.io/default-interface: 'eth0'
                networking.gke.io/interfaces: |
                  [
                    {"interfaceName":"eth0","network":"default"},
                    {"interfaceName":"eth1","network":"vpc0"},
                    {"interfaceName":"eth2","network":"vpc1"},
                    {"interfaceName":"eth3","network":"vpc2"},
                    {"interfaceName":"eth4","network":"vpc3"},
                    {"interfaceName":"eth5","network":"vpc4"},
                    {"interfaceName":"eth6","network":"vpc5"},
                    {"interfaceName":"eth7","network":"vpc6"},
                    {"interfaceName":"eth8","network":"vpc7"}
                  ]
            spec:
              activeDeadlineSeconds: 3600
              restartPolicy: Never
              nodeSelector:
                cloud.google.com/gke-accelerator: nvidia-h100-mega-80gb
              tolerations:
              - key: cloud.google.com/gke-queued
                effect: NoSchedule
                value: "true"
              - key: "nvidia.com/gpu"
                operator: "Exists"
                effect: "NoSchedule"
              setHostnameAsFQDN: true
              volumes:
              - name: nvidia
                hostPath:
                  path: /home/kubernetes/bin/nvidia
              - name: lib64
                hostPath:
                  path: /lib64
              - name: proc
                hostPath:
                  path: /proc
              - name: shared-memory
                emptyDir:
                  medium: "Memory"
                  sizeLimit: 250Gi
              - name: nccl-config
                configMap:
                  name: nccl-configmap
                  defaultMode: 0755
              containers:
              - name: nccl-test
                image: us-docker.pkg.dev/gce-ai-infra/gpudirect-tcpxo/nccl-plugin-gpudirecttcpx-dev:v1.0.15
                stdin: true
                tty: true
                securityContext:
                  privileged: true
                env:
                - name: LD_LIBRARY_PATH
                  value: /usr/local/nvidia/lib64
                volumeMounts:
                - name: nvidia
                  mountPath: /usr/local/nvidia
                - name: shared-memory
                  mountPath: /dev/shm
                - name: nccl-config
                  mountPath: /configs
                resources:
                  limits:
                    cpu: "200"
                    memory: "3700Gi"
                    nvidia.com/gpu: 8
                  requests:
                    cpu: "200"
                    memory: "3700Gi"
                    nvidia.com/gpu: 8
              - name: tcpxo-daemon
                image: us-docker.pkg.dev/gce-ai-infra/gpudirect-tcpxo/tcpgpudmarxd-dev:v1.0.21
                imagePullPolicy: Always
                command: ["/bin/sh", "-c"]
                args:
                  - |
                    set -ex
                    chmod 755 /fts/entrypoint_rxdm_container.sh
                    /fts/entrypoint_rxdm_container.sh --num_hops=2 --num_nics=8 --uid= --alsologtostderr
                securityContext:
                  privileged: true
                  capabilities:
                    add:
                      - NET_ADMIN
                      - NET_BIND_SERVICE
                volumeMounts:
                - name: nvidia
                  mountPath: /usr/local/nvidia/lib64
                - name: proc
                  mountPath: /proc
                env:
                - name: LD_LIBRARY_PATH
                  value: /usr/local/nvidia/lib64

A3 High

apiVersion: v1
kind: ConfigMap
metadata:
  name: nccl-config
data:
  allgather.sh: |
    #!/bin/bash
    for script in /configs/*; do
      name=$(basename $script)
      cp $script "/scripts/$name"
      chmod +x "/scripts/$name"
    done
    /scripts/init_ssh.sh ${@};
    pushd /scripts;
    /scripts/gen_hostfiles.sh ${@};
    popd;
    /scripts/run-allgather.sh 8 eth1,eth2,eth3,eth4 1M 512M ${#};
---
apiVersion: jobset.x-k8s.io/v1alpha2
kind: JobSet
metadata:
  name: nccl-tas-test
  labels:
    kueue.x-k8s.io/queue-name: lq-tas
spec:
  suspend: true
  network:
    enableDNSHostnames: true
  replicatedJobs:
  - name: worker
    replicas: 2
    template:
      spec:
        parallelism: 1
        completions: 1
        template:
          metadata:
            annotations:
              kueue.x-k8s.io/podset-preferred-topology: "cloud.google.com/gce-topology-block"
              networking.gke.io/default-interface: 'eth0'
              networking.gke.io/interfaces: |
                [
                  {"interfaceName":"eth0","network":"default"},
                  {"interfaceName":"eth1","network":"vpc0"},
                  {"interfaceName":"eth2","network":"vpc1"},
                  {"interfaceName":"eth3","network":"vpc2"},
                  {"interfaceName":"eth4","network":"vpc3"}
                ]
          spec:
            terminationGracePeriodSeconds: 0
            nodeSelector:
              cloud.google.com/gke-accelerator: nvidia-h100-80gb
            tolerations:
            - key: cloud.google.com/gke-queued
              effect: NoSchedule
              value: "true"
            - key: "nvidia.com/gpu"
              operator: "Exists"
              effect: "NoSchedule"
            setHostnameAsFQDN: true
            containers:
            - name: tcpx-daemon
              image: us-docker.pkg.dev/gce-ai-infra/gpudirect-tcpx/tcpgpudmarxd-dev:v2.0.11
              command:
                - /tcpgpudmarxd/build/app/tcpgpudmarxd
                - --gpu_nic_preset
                - a3vm
                - --gpu_shmem_type
                - fd
                - --uds_path
                - /run/tcpx
                - --setup_param
                - "--verbose 128 2 0 "
              securityContext:
                privileged: true
                capabilities:
                  add:
                    - NET_ADMIN
              volumeMounts:
                - name: libraries
                  mountPath: /usr/local/nvidia/lib64
                - name: tcpx-socket
                  mountPath: /run/tcpx
                - name: sys
                  mountPath: /hostsysfs
                - name: proc-sys
                  mountPath: /hostprocsysfs
              env:
                - name: LD_LIBRARY_PATH
                  value: /usr/local/nvidia/lib64
            - name: nccl-test
              image: us-docker.pkg.dev/gce-ai-infra/gpudirect-tcpx/nccl-plugin-gpudirecttcpx-dev:v3.1.8
              command:
                - bash
                - -c
                - |
                  /scripts/container_entry.sh daemon;
                  sleep infinity;
              securityContext:
                privileged: true
              volumeMounts:
                - name: tcpx-socket
                  mountPath: /tmp
                - name: libraries
                  mountPath: /usr/local/nvidia/lib64
                - name: nccl-config
                  mountPath: /configs
                - name: shared-memory
                  mountPath: /dev/shm
              resources:
                limits:
                  cpu: "200"
                  memory: "1800Gi"
                  nvidia.com/gpu: 8
                requests:
                  cpu: "200"
                  memory: "1800Gi"
                  nvidia.com/gpu: 8
            volumes:
            - name: libraries
              hostPath:
                path: /home/kubernetes/bin/nvidia/lib64
            - name: tcpx-socket
              emptyDir: {}
            - name: sys
              hostPath:
                path: /sys
            - name: proc-sys
              hostPath:
                path: /proc/sys
            - name: shared-memory
              emptyDir:
                medium: Memory
                sizeLimit: 250Gi
            - name: nccl-config
              configMap:
                name: nccl-config
                defaultMode: 0777

מחילים את המניפסט על האשכול:
```
kubectl apply -f nccl-tas-jobset.yaml
```
בודקים ש-JobSet התקבל ופועל:
```
kubectl get jobset nccl-tas-test
```
ממתינים עד שה-JobSet יבטל את ההשהיה וה-Pods יגיעו לסטטוס Running.

מפעילים את בדיקת NCCL על ידי הרצת הסקריפט allgather.sh מ-Pod העובד הראשון:

kubectl exec --stdin --tty --container=nccl-test nccl-tas-test-worker-0-0 -- /configs/allgather.sh nccl-tas-test-worker-0-0 nccl-tas-test-worker-1-0

הפלט של בדיקה עם שני צמתים אמור להיראות כך:

A3 Mega

#                                                              out-of-place                       in-place
#        size         count      type   redop    root     time   algbw   busbw #wrong     time   algbw   busbw #wrong
#         (B)    (elements)                               (us)  (GB/s)  (GB/s)            (us)  (GB/s)  (GB/s)
    0                 0         float    none      -1     0.24    0.00    0.00      0     0.18    0.00    0.00      0
    ...
    8589934592     134217728    float    none      -1    42603  201.63  189.03      0    42670  201.31  188.73      0
# Out of bounds values : 0 OK
# Avg bus bandwidth    : 45.7587

A3 High

#                                                              out-of-place                       in-place
#       size         count      type   redop    root     time   algbw   busbw #wrong     time   algbw   busbw #wrong
#        (B)    (elements)                               (us)  (GB/s)  (GB/s)            (us)  (GB/s)  (GB/s)
    1048576         16384     float    none      -1    696.8    1.50    1.41      0    729.0    1.44    1.35      0
    ...
    536870912       8388608     float    none      -1   7101.7   75.60   70.87      0   7060.9   76.03   71.28      0
# Out of bounds values : 0 OK
# Avg bus bandwidth    : 29.8293

פריסת עומס עבודה לבדיקת NCCL באמצעות TAS

אם יש לכם יותר משני צמתים, מומלץ להשתמש בבדיקה הבאה, שמשתמשת בתזמון מודע לטופולוגיה (TAS). כדי להריץ בדיקות NCCL עם TAS באשכול GKE שמשתמש במכונות וירטואליות מסוג A3 Mega או A3 High Flex-start, צריך לפעול לפי השלבים הבאים.

שומרים את קובץ המניפסט הבא בשם nccl-jobset-test.yaml. מחליפים את NUM_NODES במספר הצמתים במאגר הצמתים:

A3 Mega

apiVersion: jobset.x-k8s.io/v1alpha2
kind: JobSet
metadata:
  name: nccl-ag
  labels:
    kueue.x-k8s.io/queue-name: lq-tas
spec:
  ttlSecondsAfterFinished: 1200
  suspend: true
  network:
    enableDNSHostnames: true
  replicatedJobs:
    - name: worker
      template:
        spec:
          parallelism: NUM_NODES
          completions: NUM_NODES
          template:
            metadata:
              annotations:
                kueue.x-k8s.io/podset-preferred-topology: "cloud.google.com/gce-topology-subblock"
                networking.gke.io/default-interface: 'eth0'
                networking.gke.io/interfaces: |
                  [
                    {"interfaceName":"eth0","network":"default"},
                    {"interfaceName":"eth1","network":"vpc0"},
                    {"interfaceName":"eth2","network":"vpc1"},
                    {"interfaceName":"eth3","network":"vpc2"},
                    {"interfaceName":"eth4","network":"vpc3"},
                    {"interfaceName":"eth5","network":"vpc4"},
                    {"interfaceName":"eth6","network":"vpc5"},
                    {"interfaceName":"eth7","network":"vpc6"},
                    {"interfaceName":"eth8","network":"vpc7"}
                  ]
            spec:
              activeDeadlineSeconds: 3600
              restartPolicy: Never
              nodeSelector:
                cloud.google.com/gke-accelerator: nvidia-h100-mega-80gb
              tolerations:
              - key: "nvidia.com/gpu"
                operator: "Exists"
                effect: "NoSchedule"
              setHostnameAsFQDN: true
              volumes:
              - name: proc
                hostPath:
                  path: /proc
              - name: nvidia
                hostPath:
                  path: /home/kubernetes/bin/nvidia
              - name: lib64
                hostPath:
                  path: /lib64
              - name: shared-memory
                emptyDir:
                  medium: "Memory"
                  sizeLimit: 250Gi
              containers:
              - name: nccl-test
                stdin: true
                tty: true
                image: us-docker.pkg.dev/gce-ai-infra/gpudirect-tcpxo/nccl-plugin-tcpxo-diagnostic:v1.0.6
                securityContext:
                  privileged: true
                env:
                - name: MY_NODE_NAME
                  valueFrom:
                    fieldRef:
                      fieldPath: spec.nodeName
                - name: OMPI_ALLOW_RUN_AS_ROOT
                  value: "1"
                - name: OMPI_ALLOW_RUN_AS_ROOT_CONFIRM
                  value: "1"
                - name: N_NODES
                  value: "NUM_NODES"
                - name: NCCL_SOCKET_IFNAME
                  value: eth0
                - name: NCCL_FASTRAK_CTRL_DEV
                  value: eth0
                - name: NCCL_FASTRAK_IFNAME
                  value: eth1,eth2,eth3,eth4,eth5,eth6,eth7,eth8
                - name: NCCL_CROSS_NIC
                  value: "0"
                - name: NCCL_ALGO
                  value: Ring,Tree
                - name: NCCL_PROTO
                  value: Simple
                - name: NCCL_NET_GDR_LEVEL
                  value: PIX
                - name: LD_LIBRARY_PATH
                  value: /usr/local/nvidia/lib64
                command:
                - bash
                - -c
                - |
                  set -x
                  /scripts/container_entry.sh daemon &
                  export POSTFIX=$(hostname | cut -d . -f 2-)
                  export WORKERS_BASENAME=$(hostname | cut -d . -f 1 | rev | cut -d - -f 2- | rev )
                  export NODE_RANK=$JOB_COMPLETION_INDEX
                  for i in `seq 0 $(($N_NODES-1))`; do
                    OTHER=<span class="katex"><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.8333em;vertical-align:-0.15em;"></span><span class="mord"><span class="mord mathnormal" style="margin-right:0.13889em;">W</span><span class="mord mathnormal" style="margin-right:0.00773em;">OR</span><span class="mord mathnormal" style="margin-right:0.07153em;">K</span><span class="mord mathnormal" style="margin-right:0.00773em;">ER</span><span class="mord"><span class="mord mathnormal" style="margin-right:0.05764em;">S</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.3283em;"><span style="top:-2.55em;margin-left:-0.0576em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mathnormal mtight" style="margin-right:0.05017em;">B</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span><span class="mord mathnormal">A</span><span class="mord mathnormal" style="margin-right:0.10903em;">SEN</span><span class="mord mathnormal">A</span><span class="mord mathnormal" style="margin-right:0.05764em;">ME</span></span><span class="mord">−</span></span></span></span>{i}.${POSTFIX}
                    until ssh -p 222 -o StrictHostKeyChecking=no $OTHER hostname; do
                      sleep 10
                    done
                    echo ${OTHER} port=222 slots=8 | tee -a /tmp/hostfile;
                  done
                  if [[ "${NODE_RANK}" -eq "0" ]]; then
                      export NCCL_TESTS_SPLIT_MASK="0x0";
                      ENV_VARS=$(echo ${!NCCL*} ${!OMPI*} LD_LIBRARY_PATH PATH | sed 's/ / -x /g')
                      mpirun --hostfile /tmp/hostfile \
                        -x $ENV_VARS  \
                        -mca plm_rsh_no_tree_spawn 1 \
                        --mca orte_keep_fqdn_hostnames 1 \
                        --mca btl self,tcp \
                        --mca btl_tcp_if_include eth0 \
                        --bind-to none \
                        --mca plm_rsh_agent "ssh -q -o LogLevel=ERROR -o StrictHostKeyChecking=no -p 222" \
                        /third_party/nccl-tests/build/all_gather_perf -b 1K -e 8G -f 2 -g 1 -w 5 --iters 100 -c 1
                  else
                      while ping -c 1 <span class="katex"><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.8333em;vertical-align:-0.15em;"></span><span class="mord"><span class="mord mathnormal" style="margin-right:0.13889em;">W</span><span class="mord mathnormal" style="margin-right:0.00773em;">OR</span><span class="mord mathnormal" style="margin-right:0.07153em;">K</span><span class="mord mathnormal" style="margin-right:0.00773em;">ER</span><span class="mord"><span class="mord mathnormal" style="margin-right:0.05764em;">S</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.3283em;"><span style="top:-2.55em;margin-left:-0.0576em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mathnormal mtight" style="margin-right:0.05017em;">B</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span><span class="mord mathnormal">A</span><span class="mord mathnormal" style="margin-right:0.10903em;">SEN</span><span class="mord mathnormal">A</span><span class="mord mathnormal" style="margin-right:0.05764em;">ME</span></span><span class="mspace" style="margin-right:0.2222em;"></span><span class="mbin">−</span><span class="mspace" style="margin-right:0.2222em;"></span></span><span class="base"><span class="strut" style="height:0.6444em;"></span><span class="mord">0.</span></span></span></span>{POSTFIX}; do
                      sleep 5
                  done
                  fi
                  exit 0
                volumeMounts:
                - name: nvidia
                  mountPath: /usr/local/nvidia
                - name: lib64
                  mountPath: /lib64
                - name: shared-memory
                  mountPath: /dev/shm
                resources:
                  limits:
                    cpu: "200"
                    memory: "3700Gi"
                    nvidia.com/gpu: 8
                  requests:
                    cpu: "200"
                    memory: "3700Gi"
                    nvidia.com/gpu: 8
              - name: tcpxo-daemon
                image: us-docker.pkg.dev/gce-ai-infra/gpudirect-tcpxo/tcpxo-daemon:v1.0.1
                imagePullPolicy: Always
                command:
                - bash
                - -c
                - |
                  /usr/bin/tcpxo_daemon
                securityContext:
                  privileged: true
                volumeMounts:
                - name: nvidia
                  mountPath: /usr/local/nvidia
                - name: proc
                  mountPath: /proc
                env:
                - name: LD_LIBRARY_PATH
                  value: /usr/local/nvidia/lib64

A3 High

apiVersion: jobset.x-k8s.io/v1alpha2
kind: JobSet
metadata:
  name: nccl-ag
  labels:
    kueue.x-k8s.io/queue-name: lq-tas
spec:
  ttlSecondsAfterFinished: 1200
  suspend: true
  network:
    enableDNSHostnames: true
  replicatedJobs:
    - name: worker
      template:
        spec:
          parallelism: NUM_NODES
          completions: NUM_NODES
          template:
            metadata:
              annotations:
                kueue.x-k8s.io/podset-preferred-topology: "cloud.google.com/gce-topology-subblock"
                networking.gke.io/default-interface: 'eth0'
                networking.gke.io/interfaces: |
                  [
                    {"interfaceName":"eth0","network":"default"},
                    {"interfaceName":"eth1","network":"vpc0"},
                    {"interfaceName":"eth2","network":"vpc1"},
                    {"interfaceName":"eth3","network":"vpc2"},
                    {"interfaceName":"eth4","network":"vpc3"}
                  ]
            spec:
              activeDeadlineSeconds: 3600
              restartPolicy: Never
              nodeSelector:
                cloud.google.com/gke-accelerator: nvidia-h100-80gb
              tolerations:
              - key: "nvidia.com/gpu"
                operator: "Exists"
                effect: "NoSchedule"
              setHostnameAsFQDN: true
              volumes:
              - name: proc
                hostPath:
                  path: /proc
              - name: nvidia
                hostPath:
                  path: /home/kubernetes/bin/nvidia
              - name: libraries
                hostPath:
                  path: /home/kubernetes/bin/nvidia/lib64
              - name: tcpx-socket
                emptyDir: {}
              - name: shared-memory
                emptyDir:
                  medium: "Memory"
                  sizeLimit: 250Gi
              containers:
              - name: tcpx-daemon
                image: us-docker.pkg.dev/gce-ai-infra/gpudirect-tcpx/tcpgpudmarxd-dev:v2.0.11
                command:
                  - /tcpgpudmarxd/build/app/tcpgpudmarxd
                  - --gpu_nic_preset
                  - a3vm
                  - --uds_path
                  - /run/tcpx
                securityContext:
                  privileged: true
                volumeMounts:
                  - name: tcpx-socket
                    mountPath: /run/tcpx
                  - name: libraries
                    mountPath: /usr/local/nvidia/lib64
              - name: nccl-test
                stdin: true
                tty: true
                image: us-docker.pkg.dev/gce-ai-infra/gpudirect-tcpx/nccl-plugin-gpudirecttcpx-dev:v3.1.8
                securityContext:
                  privileged: true
                env:
                - name: MY_NODE_NAME
                  valueFrom:
                    fieldRef:
                      fieldPath: spec.nodeName
                - name: OMPI_ALLOW_RUN_AS_ROOT
                  value: "1"
                - name: OMPI_ALLOW_RUN_AS_ROOT_CONFIRM
                  value: "1"
                - name: N_NODES
                  value: "NUM_NODES"
                - name: LD_LIBRARY_PATH
                  value: /usr/local/nvidia/lib64
                command:
                - bash
                - -c
                - |
                  /scripts/container_entry.sh daemon &
                  export POSTFIX=$(hostname | cut -d . -f 2-)
                  export WORKERS_BASENAME=$(hostname | cut -d . -f 1 | rev | cut -d - -f 2- | rev )
                  export NODE_RANK=$JOB_COMPLETION_INDEX
                  for i in `seq 0 $(($N_NODES-1))`; do
                    OTHER=<span class="katex"><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.8333em;vertical-align:-0.15em;"></span><span class="mord"><span class="mord mathnormal" style="margin-right:0.13889em;">W</span><span class="mord mathnormal" style="margin-right:0.00773em;">OR</span><span class="mord mathnormal" style="margin-right:0.07153em;">K</span><span class="mord mathnormal" style="margin-right:0.00773em;">ER</span><span class="mord"><span class="mord mathnormal" style="margin-right:0.05764em;">S</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.3283em;"><span style="top:-2.55em;margin-left:-0.0576em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mathnormal mtight" style="margin-right:0.05017em;">B</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span><span class="mord mathnormal">A</span><span class="mord mathnormal" style="margin-right:0.10903em;">SEN</span><span class="mord mathnormal">A</span><span class="mord mathnormal" style="margin-right:0.05764em;">ME</span></span><span class="mord">−</span></span></span></span>{i}.${POSTFIX}
                    until ssh -p 222 -o StrictHostKeyChecking=no $OTHER hostname; do
                      sleep 10
                    done
                    echo ${OTHER} port=222 slots=8 | tee -a /tmp/hostfile;
                  done
                  if [[ "${NODE_RANK}" -eq "0" ]]; then
                      /scripts/run-allgather.sh 8 eth1,eth2,eth3,eth4 1M 512M ${N_NODES}
                  else
                      while ping -c 1 <span class="katex"><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.8333em;vertical-align:-0.15em;"></span><span class="mord"><span class="mord mathnormal" style="margin-right:0.13889em;">W</span><span class="mord mathnormal" style="margin-right:0.00773em;">OR</span><span class="mord mathnormal" style="margin-right:0.07153em;">K</span><span class="mord mathnormal" style="margin-right:0.00773em;">ER</span><span class="mord"><span class="mord mathnormal" style="margin-right:0.05764em;">S</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.3283em;"><span style="top:-2.55em;margin-left:-0.0576em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mathnormal mtight" style="margin-right:0.05017em;">B</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span><span class="mord mathnormal">A</span><span class="mord mathnormal" style="margin-right:0.10903em;">SEN</span><span class="mord mathnormal">A</span><span class="mord mathnormal" style="margin-right:0.05764em;">ME</span></span><span class="mspace" style="margin-right:0.2222em;"></span><span class="mbin">−</span><span class="mspace" style="margin-right:0.2222em;"></span></span><span class="base"><span class="strut" style="height:0.6444em;"></span><span class="mord">0.</span></span></span></span>{POSTFIX}; do
                      sleep 5
                  done
                  fi
                  exit 0
                volumeMounts:
                - name: nvidia
                  mountPath: /usr/local/nvidia
                - name: tcpx-socket
                  mountPath: /tmp
                - name: libraries
                  mountPath: /usr/local/nvidia/lib64
                - name: shared-memory
                  mountPath: /dev/shm
                resources:
                  limits:
                    cpu: "200"
                    memory: "1800Gi"
                    nvidia.com/gpu: 8
                  requests:
                    cpu: "200"
                    memory: "1800Gi"
                    nvidia.com/gpu: 8

החלת המניפסט:
```
kubectl apply -f nccl-jobset-test.yaml
```
בודקים שהעומס התקבל והגיע למצב Completed.

כדי לראות את התוצאות, מאחזרים את היומנים של ה-Pod שתואם ל-nccl-ag-worker-0-0-.*:

kubectl logs $(kubectl get pods -o go-template='{{range .items}}{{.metadata.name}}{{"\n"}}{{end}}' | grep nccl-ag-worker-0-0)

המאמרים הבאים

כדי להבין את תוצאות הבדיקה ולפתור בעיות, אפשר לעיין במאמר בנושא איסוף והבנה של יומני NCCL לצורך פתרון בעיות.
מידע נוסף על פתרון בעיות שקשורות לביצועים איטיים

הפעלת NCCL באשכולות GKE בהתאמה אישית עם A3 Mega או A3 High קל לארגן דפים בעזרת אוספים אפשר לשמור ולסווג תוכן על סמך ההעדפות שלך.

לפני שמתחילים

הגדרת האשכול באמצעות Jobset ו-Kueue

A3 High

A3 Mega

בדיקה בשני צמתים עם התחלה גמישה

A3 Mega

A3 High

A3 Mega

A3 High

פריסת עומס עבודה לבדיקת NCCL באמצעות TAS

A3 Mega

A3 High

המאמרים הבאים

הפעלת NCCL באשכולות GKE בהתאמה אישית עם A3 Mega או A3 High