Google uses AI technology to translate content into your preferred language. AI translations can contain errors.

Esegui NCCL su cluster GKE personalizzati che utilizzano A3 Mega o A3 High

Questa pagina descrive come eseguire i test della NVIDIA Collective Communications Library (NCCL) su cluster GKE personalizzati che utilizzano i protocolli di rete GPUDirect-TCPXO e GPUDirect-TCPX. Un cluster GKE personalizzato è un cluster che crei utilizzando i comandi gcloud.

Puoi utilizzare i test descritti in questa pagina per i seguenti scenari:

Se il cluster GKE utilizza nodi con avvio flessibile, esegui un test di base su due nodi.
Se il cluster GKE utilizza diversi tipi di nodi, come nodi on demand o vincolati alla prenotazione, utilizza un test NCCL con la pianificazione sensibile alla topologia.

Prima di iniziare

I test in questa pagina utilizzano JobSet e Kueue con la pianificazione sensibile alla topologia (TAS). Prima di eseguire qualsiasi test, devi configurare il cluster e procedere nel seguente modo:

Installa JobSet.

Installa Kueue.

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

Configura il cluster con JobSet e Kueue

Dopo aver installato JobSet e Kueue:

Salva il seguente manifest come 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

Applica il manifest:
```
kubectl apply -f kueue-config.yaml
```

Quando esegui i carichi di lavoro con TAS abilitato, puoi specificare la rigorosità con cui vengono applicati i vincoli di topologia utilizzando una delle seguenti annotazioni nel manifest del carico di lavoro:

kueue.x-k8s.io/podset-required-topology: se utilizzi questa annotazione, Kueue blocca la pianificazione finché il carico di lavoro non può essere pianificato all'interno del vincolo di topologia richiesto. Utilizza questa annotazione per assicurarti che i pod vengano posizionati insieme per un rendimento ottimale.
kueue.x-k8s.io/podset-preferred-topology: se utilizzi questa annotazione, Kueue tenta di pianificare i pod all'interno del vincolo di topologia richiesto, ma se non è possibile, ammette il carico di lavoro senza soddisfare i vincoli di topologia.

Nota: evita di utilizzare la modalità richiesta con DWS con avvio flessibile. Poiché l'avvio flessibile esegue il provisioning dei nodi in modo dinamico, i nodi risultanti potrebbero non soddisfare i requisiti topologici rigorosi, il che può comportare carichi di lavoro non pianificabili. Per queste configurazioni, utilizza invece podset-preferred-topology.

Per entrambe le annotazioni, specifica uno dei seguenti valori come vincolo di topologia:

cloud.google.com/gce-topology-block: pianifica i pod all'interno dello stesso blocco di rete.
cloud.google.com/gce-topology-subblock: pianifica i pod all'interno dello stesso rack.
cloud.google.com/gce-topology-host: pianifica i pod sullo stesso host fisico.

Test su due nodi con avvio flessibile

Per eseguire i test NCCL su un cluster GKE che utilizza VM con avvio flessibile A3 Mega o A3 High, segui questa procedura. Questa procedura utilizza un JobSet per eseguire un test NCCL su due nodi.

Salva il seguente manifest come 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

Applica il manifest al cluster:
```
kubectl apply -f nccl-tas-jobset.yaml
```
Verifica che JobSet sia ammesso e in esecuzione:
```
kubectl get jobset nccl-tas-test
```
Attendi che JobSet venga riattivato e che i pod raggiungano lo stato Running.

Attiva il test NCCL eseguendo lo script allgather.sh dal primo pod worker:

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

L'output di un test a due nodi è simile al seguente:

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

Esegui il deployment di un carico di lavoro di test NCCL con TAS

Se hai più di due nodi, ti consigliamo di utilizzare il seguente test, che utilizza la pianificazione sensibile alla topologia (TAS). Per eseguire i test NCCL con TAS su un cluster GKE che utilizza VM con avvio flessibile A3 Mega o A3 High, segui questa procedura.

Salva il seguente manifest come nccl-jobset-test.yaml. Sostituisci NUM_NODES con il numero di nodi nel pool di nodi:

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

Applica il manifest:
```
kubectl apply -f nccl-jobset-test.yaml
```
Verifica che il carico di lavoro sia ammesso e raggiunga lo stato Completed.

Recupera i log del pod corrispondente a nccl-ag-worker-0-0-.* per visualizzare i risultati:

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

Passaggi successivi

Raccogli e comprendi i log NCCL per la risoluzione dei problemi per comprendere gli output dei test e risolvere i problemi.
Scopri di più sulla risoluzione dei problemi di rendimento lento.

Esegui NCCL su cluster GKE personalizzati che utilizzano A3 Mega o A3 High Mantieni tutto organizzato con le raccolte Salva e classifica i contenuti in base alle tue preferenze.

Prima di iniziare

Configura il cluster con JobSet e Kueue

A3 High

A3 Mega

Test su due nodi con avvio flessibile

A3 Mega

A3 High

A3 Mega

A3 High

Esegui il deployment di un carico di lavoro di test NCCL con TAS

A3 Mega

A3 High

Passaggi successivi

Esegui NCCL su cluster GKE personalizzati che utilizzano A3 Mega o A3 High