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

Ejecuta NCCL en clústeres personalizados de GKE que usan A3 Mega o A3 High

En esta página, se describe cómo ejecutar pruebas de la biblioteca de comunicación colectiva de NVIDIA (NCCL) en clústeres personalizados de GKE que usan los protocolos de redes GPUDirect-TCPXO y GPUDirect-TCPX. Un clúster de GKE personalizado es un clúster que creas con comandos gcloud.

Puedes usar las pruebas que se describen en esta página para las siguientes situaciones:

Si tu clúster de GKE usa nodos de inicio flexible, realiza una prueba básica en dos nodos.
Si tu clúster de GKE usa diferentes tipos de nodos, como nodos bajo demanda o vinculados a reservas, usa una prueba de NCCL con Topology Aware Scheduling.

Antes de comenzar

Las pruebas en esta página usan JobSet y Kueue con la programación consciente de la topología (TAS). Antes de ejecutar cualquier prueba, debes configurar tu clúster y hacer lo siguiente:

Instala JobSet.

Instala Kueue.

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

Configura tu clúster con JobSet y Kueue

Después de instalar JobSet y Kueue, sigue estos pasos:

Guarda el siguiente manifiesto como 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

Aplica el manifiesto
```
kubectl apply -f kueue-config.yaml
```

Cuando ejecutas cargas de trabajo con TAS habilitado, puedes especificar qué tan estrictamente se aplican las restricciones de topología con una de las siguientes anotaciones en el manifiesto de tu carga de trabajo:

kueue.x-k8s.io/podset-required-topology: Si usas esta anotación, Kueue bloquea la programación hasta que la carga de trabajo se pueda programar dentro de la restricción de topología solicitada. Usa esta anotación para asegurarte de que los pods se coloquen juntos y obtener un rendimiento óptimo.
kueue.x-k8s.io/podset-preferred-topology: Si usas esta anotación, Kueue intenta programar Pods dentro de la restricción de topología solicitada, pero, si eso no es posible, admite la carga de trabajo sin cumplir con las restricciones de topología.

Nota: Evita usar el modo obligatorio con DWS inicio flexible. Dado que el inicio flexible aprovisiona nodos de forma dinámica, es posible que los nodos resultantes no satisfagan requisitos topológicos estrictos, lo que puede generar cargas de trabajo no programables. Para estas configuraciones, usa podset-preferred-topology en su lugar.

Para cualquiera de las anotaciones, especifica uno de los siguientes valores como la restricción de topología:

cloud.google.com/gce-topology-block: Programa Pods dentro del mismo bloque de red.
cloud.google.com/gce-topology-subblock: Programa pods dentro del mismo rack.
cloud.google.com/gce-topology-host: Programa pods en el mismo host físico.

Prueba en dos nodos de inicio flexible

Para ejecutar pruebas de NCCL en un clúster de GKE que usa VMs A3 Mega o A3 High Flex-start, sigue este procedimiento. En este procedimiento, se usa un manifiesto de JobSet para ejecutar una prueba de NCCL en dos nodos.

Guarda el siguiente manifiesto como 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

Aplica el manifiesto al clúster:
```
kubectl apply -f nccl-tas-jobset.yaml
```
Verifica que el JobSet se admita y se ejecute:
```
kubectl get jobset nccl-tas-test
```
Espera a que se reanude el JobSet y a que los Pods alcancen el estado Running.

Para activar la prueba de NCCL, ejecuta la secuencia de comandos allgather.sh desde el primer Pod de trabajador:

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

El resultado de una prueba de dos nodos es similar al siguiente:

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

Implementa una carga de trabajo de prueba de NCCL con TAS

Si tienes más de dos nodos, te recomendamos que uses la siguiente prueba, que utiliza la programación con reconocimiento de la topología (TAS). Para ejecutar pruebas de NCCL con TAS en un clúster de GKE que usa VMs A3 Mega o A3 High con inicio flexible, sigue este procedimiento.

Guarda el siguiente manifiesto como nccl-jobset-test.yaml. Reemplaza NUM_NODES por la cantidad de nodos del grupo de nodos:

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

Aplica el manifiesto
```
kubectl apply -f nccl-jobset-test.yaml
```
Comprueba que se admita la carga de trabajo y que alcance el estado Completed.

Recupera los registros del Pod que coincide con nccl-ag-worker-0-0-.* para ver los resultados:

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

¿Qué sigue?

Collect and Understand NCCL Logs for Troubleshooting para comprender los resultados de las pruebas y solucionar problemas
Obtén más información para solucionar problemas de rendimiento lento.

Ejecuta NCCL en clústeres personalizados de GKE que usan A3 Mega o A3 High Organiza tus páginas con colecciones Guarda y categoriza el contenido según tus preferencias.

Antes de comenzar

Configura tu clúster con JobSet y Kueue

A3 High

A3 Mega

Prueba en dos nodos de inicio flexible

A3 Mega

A3 High

A3 Mega

A3 High

Implementa una carga de trabajo de prueba de NCCL con TAS

A3 Mega

A3 High

¿Qué sigue?

Ejecuta NCCL en clústeres personalizados de GKE que usan A3 Mega o A3 High