Accueiil
Products
OpenShift Container Platform
4.18
Hardware accelerators
Chapter 5. NVIDIA GPUDirect Remote Direct Memory Access (RDMA)

Ce contenu n'est pas disponible dans la langue sélectionnée.

Chapter 5. NVIDIA GPUDirect Remote Direct Memory Access (RDMA)

NVIDIA GPUDirect Remote Direct Memory Access (RDMA) allows for an application in one computer to directly access the memory of another computer without needing access through the operating system. This provides the ability to bypass kernel intervention in the process, freeing up resources and greatly reducing the CPU overhead normally needed to process network communications. This is useful for distributing GPU-accelerated workloads across clusters. And because RDMA is so suited toward high bandwidth and low latency applications, this makes it ideal for big data and machine learning applications.

There are currently three configuration methods for NVIDIA GPUDirect RDMA:

Shared device: This method allows for an NVIDIA GPUDirect RDMA device to be shared among multiple pods on the OpenShift Container Platform worker node where the device is exposed.
Host device: This method provides direct physical Ethernet access on the worker node by creating an additional host network on a pod. A plugin allows the network device to be moved from the host network namespace to the network namespace on the pod.
SR-IOV legacy device: The Single Root IO Virtualization (SR-IOV) method can share a single network device, such as an Ethernet adapter, with multiple pods. SR-IOV segments the device, recognized on the host node as a physical function (PF), into multiple virtual functions (VFs). The VF is used like any other network device.

Each of these methods can be used across either the NVIDIA GPUDirect RDMA over Converged Ethernet (RoCE) or Infiniband infrastructures, providing an aggregate total of six methods of configuration.

5.1. NVIDIA GPUDirect RDMA prerequisites
Copier lien

All methods of NVIDIA GPUDirect RDMA configuration require the installation of specific Operators. Use the following steps to install the Operators:

Install the Node Feature Discovery Operator.
Install the SR-IOV Operator.
Install the NVIDIA Network Operator (NVIDIA documentation).
Install the NVIDIA GPU Operator (NVIDIA documentation).

5.2. Disabling the IRDMA kernel module
Copier lien

On some systems, including the DellR750xa, the IRDMA kernel module creates problems for the NVIDIA Network Operator when unloading and loading the DOCA drivers. Use the following procedure to disable the module.

Procedure

Generate the following machine configuration file by running the following command:

$ cat <<EOF > 99-machine-config-blacklist-irdma.yaml

Example output

apiVersion: machineconfiguration.openshift.io/v1
kind: MachineConfig
metadata:
  labels:
    machineconfiguration.openshift.io/role: worker
  name: 99-worker-blacklist-irdma
spec:
  kernelArguments:
    - "module_blacklist=irdma"

Create the machine configuration on the cluster and wait for the nodes to reboot by running the following command:

$ oc create -f 99-machine-config-blacklist-irdma.yaml

Example output

machineconfig.machineconfiguration.openshift.io/99-worker-blacklist-irdma created

Validate in a debug pod on each node that the module has not loaded by running the following command:

$ oc debug node/nvd-srv-32.nvidia.eng.rdu2.dc.redhat.com
Starting pod/nvd-srv-32nvidiaengrdu2dcredhatcom-debug-btfj2 ...
To use host binaries, run `chroot /host`
Pod IP: 10.6.135.11
If you don't see a command prompt, try pressing enter.
sh-5.1# chroot /host
sh-5.1# lsmod|grep irdma
sh-5.1#

5.3. Creating persistent naming rules
Copier lien

In some cases, device names won’t persist following a reboot. For example, on R760xa systems Mellanox devices might be renamed after a reboot. You can avoid this problem by using a MachineConfig to set persistence.

Procedure

Gather the MAC address names from the worker nodes for the node into a file and provide names for the interfaces that need to persist. This example uses the file 70-persistent-net.rules and stashes the details in it.

$ cat <<EOF > 70-persistent-net.rules
SUBSYSTEM=="net",ACTION=="add",ATTR{address}=="b8:3f:d2:3b:51:28",ATTR{type}=="1",NAME="ibs2f0"
SUBSYSTEM=="net",ACTION=="add",ATTR{address}=="b8:3f:d2:3b:51:29",ATTR{type}=="1",NAME="ens8f0np0"
SUBSYSTEM=="net",ACTION=="add",ATTR{address}=="b8:3f:d2:f0:36:d0",ATTR{type}=="1",NAME="ibs2f0"
SUBSYSTEM=="net",ACTION=="add",ATTR{address}=="b8:3f:d2:f0:36:d1",ATTR{type}=="1",NAME="ens8f0np0"
EOF

Convert that file into a base64 string without line breaks and set the output to the variable PERSIST:

$ PERSIST=`cat 70-persistent-net.rules| base64 -w 0`

$ echo $PERSIST

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

Create a machine configuration and set the base64 encoding in the custom resource file by running the following command:

$ cat <<EOF > 99-machine-config-udev-network.yaml

apiVersion: machineconfiguration.openshift.io/v1
kind: MachineConfig
metadata:
   labels:
     machineconfiguration.openshift.io/role: worker
   name: 99-machine-config-udev-network
spec:
   config:
     ignition:
       version: 3.2.0
     storage:
       files:
       - contents:
           source: data:text/plain;base64,$PERSIST
         filesystem: root
         mode: 420
         path: /etc/udev/rules.d/70-persistent-net.rules

Create the machine configuration on the cluster by running the following command. After running the command, the expected output shows machineconfig.machineconfiguration.openshift.io/99-machine-config-udev-network created.
```
$ oc create -f 99-machine-config-udev-network.yaml
```

Use the get mcp command to view the machine configuration status:

$ oc get mcp

Example output

NAME     CONFIG                                             UPDATED   UPDATING   DEGRADED   MACHINECOUNT   READYMACHINECOUNT   UPDATEDMACHINECOUNT   DEGRADEDMACHINECOUNT   AGE
master   rendered-master-9adfe851c2c14d9598eea5ec3df6c187   True      False      False      1              1                   1                     0                      6h21m
worker   rendered-worker-4568f1b174066b4b1a4de794cf538fee   False     True       False      2              0                   0                     0                      6h21m

The nodes will reboot and when the updating field returns to false, you can validate on the nodes by looking at the devices in a debug pod.

5.4. Configuring the NFD Operator
Copier lien

The Node Feature Discovery (NFD) Operator manages the detection of hardware features and configuration in an OpenShift Container Platform cluster by labeling the nodes with hardware-specific information. NFD labels the host with node-specific attributes, such as PCI cards, kernel, operating system version, and so on.

Prerequisites

You have installed the NFD Operator.

Procedure

Validate that the Operator is installed and running by looking at the pods in the openshift-nfd namespace by running the following command:

$ oc get pods -n openshift-nfd

Example output

NAME                                      READY   STATUS    RESTARTS   AGE
nfd-controller-manager-8698c88cdd-t8gbc   2/2     Running   0          2m

With the NFD controller running, generate the NodeFeatureDiscovery instance and add it to the cluster.
The ClusterServiceVersion specification for NFD Operator provides default values, including the NFD operand image that is part of the Operator payload. Retrieve its value by running the following command:
```
$ NFD_OPERAND_IMAGE=`echo $(oc get csv -n openshift-nfd -o json | jq -r '.items[0].metadata.annotations["alm-examples"]') | jq -r '.[] | select(.kind == "NodeFeatureDiscovery") | .spec.operand.image'`
```

Optional: Add entries to the default deviceClassWhiteList field, to support more network adapters, such as the NVIDIA BlueField DPUs.

apiVersion: nfd.openshift.io/v1
kind: NodeFeatureDiscovery
metadata:
  name: nfd-instance
  namespace: openshift-nfd
spec:
  instance: ''
  operand:
    image: '${NFD_OPERAND_IMAGE}'
    servicePort: 12000
  prunerOnDelete: false
  topologyUpdater: false
  workerConfig:
    configData: |
      core:
        sleepInterval: 60s
      sources:
        pci:
          deviceClassWhitelist:
            - "02"
            - "03"
            - "0200"
            - "0207"
            - "12"
          deviceLabelFields:
            - "vendor"

Create the 'NodeFeatureDiscovery` instance by running the following command:

$ oc create -f nfd-instance.yaml

Example output

nodefeaturediscovery.nfd.openshift.io/nfd-instance created

Validate that the instance is up and running by looking at the pods under the openshift-nfd namespace by running the following command:

$ oc get pods -n openshift-nfd

Example output

NAME                                    READY   STATUS    RESTARTS   AGE
nfd-controller-manager-7cb6d656-jcnqb   2/2     Running   0          4m
nfd-gc-7576d64889-s28k9                 1/1     Running   0          21s
nfd-master-b7bcf5cfd-qnrmz              1/1     Running   0          21s
nfd-worker-96pfh                        1/1     Running   0          21s
nfd-worker-b2gkg                        1/1     Running   0          21s
nfd-worker-bd9bk                        1/1     Running   0          21s
nfd-worker-cswf4                        1/1     Running   0          21s
nfd-worker-kp6gg                        1/1     Running   0          21s

Wait a short period of time and then verify that NFD has added labels to the node. The NFD labels are prefixed with feature.node.kubernetes.io, so you can easily filter them.

$ oc get node -o json | jq '.items[0].metadata.labels | with_entries(select(.key | startswith("feature.node.kubernetes.io")))'
{
  "feature.node.kubernetes.io/cpu-cpuid.ADX": "true",
  "feature.node.kubernetes.io/cpu-cpuid.AESNI": "true",
  "feature.node.kubernetes.io/cpu-cpuid.AVX": "true",
  "feature.node.kubernetes.io/cpu-cpuid.AVX2": "true",
  "feature.node.kubernetes.io/cpu-cpuid.CETSS": "true",
  "feature.node.kubernetes.io/cpu-cpuid.CLZERO": "true",
  "feature.node.kubernetes.io/cpu-cpuid.CMPXCHG8": "true",
  "feature.node.kubernetes.io/cpu-cpuid.CPBOOST": "true",
  "feature.node.kubernetes.io/cpu-cpuid.EFER_LMSLE_UNS": "true",
  "feature.node.kubernetes.io/cpu-cpuid.FMA3": "true",
  "feature.node.kubernetes.io/cpu-cpuid.FP256": "true",
  "feature.node.kubernetes.io/cpu-cpuid.FSRM": "true",
  "feature.node.kubernetes.io/cpu-cpuid.FXSR": "true",
  "feature.node.kubernetes.io/cpu-cpuid.FXSROPT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBPB": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBRS": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBRS_PREFERRED": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBRS_PROVIDES_SMP": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBS": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBSBRNTRGT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBSFETCHSAM": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBSFFV": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBSOPCNT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBSOPCNTEXT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBSOPSAM": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBSRDWROPCNT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBSRIPINVALIDCHK": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBS_FETCH_CTLX": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBS_OPFUSE": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IBS_PREVENTHOST": "true",
  "feature.node.kubernetes.io/cpu-cpuid.INT_WBINVD": "true",
  "feature.node.kubernetes.io/cpu-cpuid.INVLPGB": "true",
  "feature.node.kubernetes.io/cpu-cpuid.LAHF": "true",
  "feature.node.kubernetes.io/cpu-cpuid.LBRVIRT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.MCAOVERFLOW": "true",
  "feature.node.kubernetes.io/cpu-cpuid.MCOMMIT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.MOVBE": "true",
  "feature.node.kubernetes.io/cpu-cpuid.MOVU": "true",
  "feature.node.kubernetes.io/cpu-cpuid.MSRIRC": "true",
  "feature.node.kubernetes.io/cpu-cpuid.MSR_PAGEFLUSH": "true",
  "feature.node.kubernetes.io/cpu-cpuid.NRIPS": "true",
  "feature.node.kubernetes.io/cpu-cpuid.OSXSAVE": "true",
  "feature.node.kubernetes.io/cpu-cpuid.PPIN": "true",
  "feature.node.kubernetes.io/cpu-cpuid.PSFD": "true",
  "feature.node.kubernetes.io/cpu-cpuid.RDPRU": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SEV": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SEV_64BIT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SEV_ALTERNATIVE": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SEV_DEBUGSWAP": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SEV_ES": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SEV_RESTRICTED": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SEV_SNP": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SHA": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SME": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SME_COHERENT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SPEC_CTRL_SSBD": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SSE4A": "true",
  "feature.node.kubernetes.io/cpu-cpuid.STIBP": "true",
  "feature.node.kubernetes.io/cpu-cpuid.STIBP_ALWAYSON": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SUCCOR": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SVM": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SVMDA": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SVMFBASID": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SVML": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SVMNP": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SVMPF": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SVMPFT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SYSCALL": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SYSEE": "true",
  "feature.node.kubernetes.io/cpu-cpuid.TLB_FLUSH_NESTED": "true",
  "feature.node.kubernetes.io/cpu-cpuid.TOPEXT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.TSCRATEMSR": "true",
  "feature.node.kubernetes.io/cpu-cpuid.VAES": "true",
  "feature.node.kubernetes.io/cpu-cpuid.VMCBCLEAN": "true",
  "feature.node.kubernetes.io/cpu-cpuid.VMPL": "true",
  "feature.node.kubernetes.io/cpu-cpuid.VMSA_REGPROT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.VPCLMULQDQ": "true",
  "feature.node.kubernetes.io/cpu-cpuid.VTE": "true",
  "feature.node.kubernetes.io/cpu-cpuid.WBNOINVD": "true",
  "feature.node.kubernetes.io/cpu-cpuid.X87": "true",
  "feature.node.kubernetes.io/cpu-cpuid.XGETBV1": "true",
  "feature.node.kubernetes.io/cpu-cpuid.XSAVE": "true",
  "feature.node.kubernetes.io/cpu-cpuid.XSAVEC": "true",
  "feature.node.kubernetes.io/cpu-cpuid.XSAVEOPT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.XSAVES": "true",
  "feature.node.kubernetes.io/cpu-hardware_multithreading": "false",
  "feature.node.kubernetes.io/cpu-model.family": "25",
  "feature.node.kubernetes.io/cpu-model.id": "1",
  "feature.node.kubernetes.io/cpu-model.vendor_id": "AMD",
  "feature.node.kubernetes.io/kernel-config.NO_HZ": "true",
  "feature.node.kubernetes.io/kernel-config.NO_HZ_FULL": "true",
  "feature.node.kubernetes.io/kernel-selinux.enabled": "true",
  "feature.node.kubernetes.io/kernel-version.full": "5.14.0-427.35.1.el9_4.x86_64",
  "feature.node.kubernetes.io/kernel-version.major": "5",
  "feature.node.kubernetes.io/kernel-version.minor": "14",
  "feature.node.kubernetes.io/kernel-version.revision": "0",
  "feature.node.kubernetes.io/memory-numa": "true",
  "feature.node.kubernetes.io/network-sriov.capable": "true",
  "feature.node.kubernetes.io/pci-102b.present": "true",
  "feature.node.kubernetes.io/pci-10de.present": "true",
  "feature.node.kubernetes.io/pci-10de.sriov.capable": "true",
  "feature.node.kubernetes.io/pci-15b3.present": "true",
  "feature.node.kubernetes.io/pci-15b3.sriov.capable": "true",
  "feature.node.kubernetes.io/rdma.available": "true",
  "feature.node.kubernetes.io/rdma.capable": "true",
  "feature.node.kubernetes.io/storage-nonrotationaldisk": "true",
  "feature.node.kubernetes.io/system-os_release.ID": "rhcos",
  "feature.node.kubernetes.io/system-os_release.OPENSHIFT_VERSION": "4.17",
  "feature.node.kubernetes.io/system-os_release.OSTREE_VERSION": "417.94.202409121747-0",
  "feature.node.kubernetes.io/system-os_release.RHEL_VERSION": "9.4",
  "feature.node.kubernetes.io/system-os_release.VERSION_ID": "4.17",
  "feature.node.kubernetes.io/system-os_release.VERSION_ID.major": "4",
  "feature.node.kubernetes.io/system-os_release.VERSION_ID.minor": "17"
}

Confirm there is a network device that is discovered:

$ oc describe node | grep -E 'Roles|pci' | grep pci-15b3
                    feature.node.kubernetes.io/pci-15b3.present=true
                    feature.node.kubernetes.io/pci-15b3.sriov.capable=true
                    feature.node.kubernetes.io/pci-15b3.present=true
                    feature.node.kubernetes.io/pci-15b3.sriov.capable=true

5.5. Configuring the SR-IOV Operator
Copier lien

Single root I/O virtualization (SR-IOV) enhances the performance of NVIDIA GPUDirect RDMA by providing sharing across multiple pods from a single device.

Prerequisites

You have installed the SR-IOV Operator.

Procedure

Validate that the Operator is installed and running by looking at the pods in the openshift-sriov-network-operator namespace by running the following command:

$ oc get pods -n openshift-sriov-network-operator

Example output

NAME                                      READY   STATUS    RESTARTS   AGE
sriov-network-operator-7cb6c49868-89486   1/1     Running   0          22s

For the default SriovOperatorConfig CR to work with the MLNX_OFED container, run this command to update the following values:

apiVersion: sriovnetwork.openshift.io/v1
kind: SriovOperatorConfig
metadata:
  name: default
  namespace: openshift-sriov-network-operator
spec:
  enableInjector: true
  enableOperatorWebhook: true
  logLevel: 2

Create the resource on the cluster by running the following command:

$ oc create -f sriov-operator-config.yaml

Example output

sriovoperatorconfig.sriovnetwork.openshift.io/default created

Patch the sriov-operator so the MOFED container can work with it by running the following command:

$ oc patch sriovoperatorconfig default   --type=merge -n openshift-sriov-network-operator   --patch '{ "spec": { "configDaemonNodeSelector": { "network.nvidia.com/operator.mofed.wait": "false", "node-role.kubernetes.io/worker": "", "feature.node.kubernetes.io/pci-15b3.sriov.capable": "true" } } }'

Example output

sriovoperatorconfig.sriovnetwork.openshift.io/default patched

5.6. Configuring the NVIDIA network Operator
Copier lien

The NVIDIA network Operator manages NVIDIA networking resources and networking related components such as drivers and device plugins to enable NVIDIA GPUDirect RDMA workloads.

Prerequisites

You have installed the NVIDIA network Operator.

Procedure

Validate that the network Operator is installed and running by confirming the controller is running in the nvidia-network-operator namespace by running the following command:

$ oc get pods -n nvidia-network-operator

Example output

NAME                                                          READY   STATUS             RESTARTS        AGE
nvidia-network-operator-controller-manager-6f7d6956cd-fw5wg   1/1     Running            0                5m

With the Operator running, create the NicClusterPolicy custom resource file. The device you choose depends on your system configuration. In this example, the Infiniband interface ibs2f0 is hard coded and is used as the shared NVIDIA GPUDirect RDMA device.

apiVersion: mellanox.com/v1alpha1
kind: NicClusterPolicy
metadata:
  name: nic-cluster-policy
spec:
  nicFeatureDiscovery:
    image: nic-feature-discovery
    repository: ghcr.io/mellanox
    version: v0.0.1
  docaTelemetryService:
    image: doca_telemetry
    repository: nvcr.io/nvidia/doca
    version: 1.16.5-doca2.6.0-host
  rdmaSharedDevicePlugin:
    config: |
      {
        "configList": [
          {
            "resourceName": "rdma_shared_device_ib",
            "rdmaHcaMax": 63,
            "selectors": {
              "ifNames": ["ibs2f0"]
            }
          },
          {
            "resourceName": "rdma_shared_device_eth",
            "rdmaHcaMax": 63,
            "selectors": {
              "ifNames": ["ens8f0np0"]
            }
          }
        ]
      }
    image: k8s-rdma-shared-dev-plugin
    repository: ghcr.io/mellanox
    version: v1.5.1
  secondaryNetwork:
    ipoib:
      image: ipoib-cni
      repository: ghcr.io/mellanox
      version: v1.2.0
  nvIpam:
    enableWebhook: false
    image: nvidia-k8s-ipam
    repository: ghcr.io/mellanox
    version: v0.2.0
  ofedDriver:
    readinessProbe:
      initialDelaySeconds: 10
      periodSeconds: 30
    forcePrecompiled: false
    terminationGracePeriodSeconds: 300
    livenessProbe:
      initialDelaySeconds: 30
      periodSeconds: 30
    upgradePolicy:
      autoUpgrade: true
      drain:
        deleteEmptyDir: true
        enable: true
        force: true
        timeoutSeconds: 300
        podSelector: ''
      maxParallelUpgrades: 1
      safeLoad: false
      waitForCompletion:
        timeoutSeconds: 0
    startupProbe:
      initialDelaySeconds: 10
      periodSeconds: 20
    image: doca-driver
    repository: nvcr.io/nvidia/mellanox
    version: 24.10-0.7.0.0-0
    env:
    - name: UNLOAD_STORAGE_MODULES
      value: "true"
    - name: RESTORE_DRIVER_ON_POD_TERMINATION
      value: "true"
    - name: CREATE_IFNAMES_UDEV
      value: "true"

Create the NicClusterPolicy custom resource on the cluster by running the following command:

$ oc create -f network-sharedrdma-nic-cluster-policy.yaml

Example output

nicclusterpolicy.mellanox.com/nic-cluster-policy created

Validate the NicClusterPolicy by running the following command in the DOCA/MOFED container:

$ oc get pods -n nvidia-network-operator

Example output

NAME                                                          READY   STATUS    RESTARTS   AGE
doca-telemetry-service-hwj65                                  1/1     Running   2          160m
kube-ipoib-cni-ds-fsn8g                                       1/1     Running   2          160m
mofed-rhcos4.16-9b5ddf4c6-ds-ct2h5                            2/2     Running   4          160m
nic-feature-discovery-ds-dtksz                                1/1     Running   2          160m
nv-ipam-controller-854585f594-c5jpp                           1/1     Running   2          160m
nv-ipam-controller-854585f594-xrnp5                           1/1     Running   2          160m
nv-ipam-node-xqttl                                            1/1     Running   2          160m
nvidia-network-operator-controller-manager-5798b564cd-5cq99   1/1     Running   2         5d23h
rdma-shared-dp-ds-p9vvg                                       1/1     Running   0          85m

rsh into the mofed container to check the status by running the following command:

$ MOFED_POD=$(oc get pods -n nvidia-network-operator -o name | grep mofed)
$ oc rsh -n nvidia-network-operator -c mofed-container ${MOFED_POD}
sh-5.1# ofed_info -s

Example output

OFED-internal-24.07-0.6.1:

sh-5.1# ibdev2netdev -v

Example output

0000:0d:00.0 mlx5_0 (MT41692 - 900-9D3B4-00EN-EA0) BlueField-3 E-series SuperNIC 400GbE/NDR single port QSFP112, PCIe Gen5.0 x16 FHHL, Crypto Enabled, 16GB DDR5, BMC, Tall Bracket                                                       fw 32.42.1000 port 1 (ACTIVE) ==> ibs2f0 (Up)
0000:a0:00.0 mlx5_1 (MT41692 - 900-9D3B4-00EN-EA0) BlueField-3 E-series SuperNIC 400GbE/NDR single port QSFP112, PCIe Gen5.0 x16 FHHL, Crypto Enabled, 16GB DDR5, BMC, Tall Bracket                                                       fw 32.42.1000 port 1 (ACTIVE) ==> ens8f0np0 (Up)

Create a IPoIBNetwork custom resource file:

apiVersion: mellanox.com/v1alpha1
kind: IPoIBNetwork
metadata:
  name: example-ipoibnetwork
spec:
  ipam: |
    {
      "type": "whereabouts",
      "range": "192.168.6.225/28",
      "exclude": [
       "192.168.6.229/30",
       "192.168.6.236/32"
      ]
    }
  master: ibs2f0
  networkNamespace: default

Create the IPoIBNetwork resource on the cluster by running the following command:
```
$ oc create -f ipoib-network.yaml
```
Example output
```
ipoibnetwork.mellanox.com/example-ipoibnetwork created
```

Create a MacvlanNetwork custom resource file for your other interface:

apiVersion: mellanox.com/v1alpha1
kind: MacvlanNetwork
metadata:
  name: rdmashared-net
spec:
  networkNamespace: default
  master: ens8f0np0
  mode: bridge
  mtu: 1500
  ipam: '{"type": "whereabouts", "range": "192.168.2.0/24", "gateway": "192.168.2.1"}'

Create the resource on the cluster by running the following command:

$ oc create -f macvlan-network.yaml

Example output

macvlannetwork.mellanox.com/rdmashared-net created

5.7. Configuring the GPU Operator
Copier lien

The GPU Operator automates the management of the NVIDIA drivers, device plugins for GPUs, the NVIDIA Container Toolkit, and other components required for GPU provisioning.

Prerequisites

You have installed the GPU Operator.

Procedure

Check that the Operator pod is running to look at the pods under the namespace by running the following command:

$ oc get pods -n nvidia-gpu-operator

Example output

NAME                          READY   STATUS    RESTARTS   AGE
gpu-operator-b4cb7d74-zxpwq   1/1     Running   0          32s

Create a GPU cluster policy custom resource file similar to the following example:

apiVersion: nvidia.com/v1
kind: ClusterPolicy
metadata:
  name: gpu-cluster-policy
spec:
  vgpuDeviceManager:
    config:
      default: default
    enabled: true
  migManager:
    config:
      default: all-disabled
      name: default-mig-parted-config
    enabled: true
  operator:
    defaultRuntime: crio
    initContainer: {}
    runtimeClass: nvidia
    use_ocp_driver_toolkit: true
  dcgm:
    enabled: true
  gfd:
    enabled: true
  dcgmExporter:
    config:
      name: ''
    serviceMonitor:
      enabled: true
    enabled: true
  cdi:
    default: false
    enabled: false
  driver:
    licensingConfig:
      nlsEnabled: true
      configMapName: ''
    certConfig:
      name: ''
    rdma:
      enabled: false
    kernelModuleConfig:
      name: ''
    upgradePolicy:
      autoUpgrade: true
      drain:
        deleteEmptyDir: false
        enable: false
        force: false
        timeoutSeconds: 300
      maxParallelUpgrades: 1
      maxUnavailable: 25%
      podDeletion:
        deleteEmptyDir: false
        force: false
        timeoutSeconds: 300
      waitForCompletion:
        timeoutSeconds: 0
    repoConfig:
      configMapName: ''
    virtualTopology:
      config: ''
    enabled: true
    useNvidiaDriverCRD: false
    useOpenKernelModules: true
  devicePlugin:
    config:
      name: ''
      default: ''
    mps:
      root: /run/nvidia/mps
    enabled: true
  gdrcopy:
    enabled: true
  kataManager:
    config:
      artifactsDir: /opt/nvidia-gpu-operator/artifacts/runtimeclasses
  mig:
    strategy: single
  sandboxDevicePlugin:
    enabled: true
  validator:
    plugin:
      env:
        - name: WITH_WORKLOAD
          value: 'false'
  nodeStatusExporter:
    enabled: true
  daemonsets:
    rollingUpdate:
      maxUnavailable: '1'
    updateStrategy: RollingUpdate
  sandboxWorkloads:
    defaultWorkload: container
    enabled: false
  gds:
    enabled: true
    image: nvidia-fs
    version: 2.20.5
    repository: nvcr.io/nvidia/cloud-native
  vgpuManager:
    enabled: false
  vfioManager:
    enabled: true
  toolkit:
    installDir: /usr/local/nvidia
    enabled: true

When the GPU ClusterPolicy custom resource has generated, create the resource on the cluster by running the following command:
```
$ oc create -f gpu-cluster-policy.yaml
```
Example output
```
clusterpolicy.nvidia.com/gpu-cluster-policy created
```

Validate that the Operator is installed and running by running the following command:

$ oc get pods -n nvidia-gpu-operator

Example output

NAME                                                  READY   STATUS      RESTARTS   AGE
gpu-feature-discovery-d5ngn                           1/1     Running     0          3m20s
gpu-feature-discovery-z42rx                           1/1     Running     0          3m23s
gpu-operator-6bb4d4b4c5-njh78                         1/1     Running     0          4m35s
nvidia-container-toolkit-daemonset-bkh8l              1/1     Running     0          3m20s
nvidia-container-toolkit-daemonset-c4hzm              1/1     Running     0          3m23s
nvidia-cuda-validator-4blvg                           0/1     Completed   0          106s
nvidia-cuda-validator-tw8sl                           0/1     Completed   0          112s
nvidia-dcgm-exporter-rrw4g                            1/1     Running     0          3m20s
nvidia-dcgm-exporter-xc78t                            1/1     Running     0          3m23s
nvidia-dcgm-nvxpf                                     1/1     Running     0          3m20s
nvidia-dcgm-snj4j                                     1/1     Running     0          3m23s
nvidia-device-plugin-daemonset-fk2xz                  1/1     Running     0          3m23s
nvidia-device-plugin-daemonset-wq87j                  1/1     Running     0          3m20s
nvidia-driver-daemonset-416.94.202410211619-0-ngrjg   4/4     Running     0          3m58s
nvidia-driver-daemonset-416.94.202410211619-0-tm4x6   4/4     Running     0          3m58s
nvidia-node-status-exporter-jlzxh                     1/1     Running     0          3m57s
nvidia-node-status-exporter-zjffs                     1/1     Running     0          3m57s
nvidia-operator-validator-l49hx                       1/1     Running     0          3m20s
nvidia-operator-validator-n44nn                       1/1     Running     0          3m23s

Optional: When you have verified the pods are running, remote shell into the NVIDIA driver daemonset pod and confirm that the NVIDIA modules are loaded. Specifically, ensure the nvidia_peermem is loaded.

$ oc rsh -n nvidia-gpu-operator $(oc -n nvidia-gpu-operator get pod -o name -l app.kubernetes.io/component=nvidia-driver)
sh-4.4# lsmod|grep nvidia

Example output

nvidia_fs             327680  0
nvidia_peermem         24576  0
nvidia_modeset       1507328  0
video                  73728  1 nvidia_modeset
nvidia_uvm           6889472  8
nvidia               8810496  43 nvidia_uvm,nvidia_peermem,nvidia_fs,gdrdrv,nvidia_modeset
ib_uverbs             217088  3 nvidia_peermem,rdma_ucm,mlx5_ib
drm                   741376  5 drm_kms_helper,drm_shmem_helper,nvidia,mgag200

Optional: Run the nvidia-smi utility to show the details about the driver and the hardware:

sh-4.4# nvidia-smi

+ .Example output

Wed Nov  6 22:03:53 2024
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 550.90.07              Driver Version: 550.90.07      CUDA Version: 12.4     |
|-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA A40                     On  |   00000000:61:00.0 Off |                    0 |
|  0%   37C    P0             88W /  300W |       1MiB /  46068MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+
|   1  NVIDIA A40                     On  |   00000000:E1:00.0 Off |                    0 |
|  0%   28C    P8             29W /  300W |       1MiB /  46068MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI        PID   Type   Process name                              GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|  No running processes found                                                             |
+-----------------------------------------------------------------------------------------+

While you are still in the driver pod, set the GPU clock to maximum using the nvidia-smi command:

$ oc rsh -n nvidia-gpu-operator nvidia-driver-daemonset-416.94.202410172137-0-ndhzc
sh-4.4# nvidia-smi -i 0 -lgc $(nvidia-smi -i 0 --query-supported-clocks=graphics --format=csv,noheader,nounits | sort -h | tail -n 1)

Example output

GPU clocks set to "(gpuClkMin 1740, gpuClkMax 1740)" for GPU 00000000:61:00.0
All done.

sh-4.4# nvidia-smi -i 1 -lgc $(nvidia-smi -i 1 --query-supported-clocks=graphics --format=csv,noheader,nounits | sort -h | tail -n 1)

Example output

GPU clocks set to "(gpuClkMin 1740, gpuClkMax 1740)" for GPU 00000000:E1:00.0
All done.

Validate the resource is available from a node describe perspective by running the following command:

$ oc describe node -l node-role.kubernetes.io/worker=| grep -E 'Capacity:|Allocatable:' -A9

Example output

Capacity:
  cpu:                          128
  ephemeral-storage:            1561525616Ki
  hugepages-1Gi:                0
  hugepages-2Mi:                0
  memory:                       263596712Ki
  nvidia.com/gpu:               2
  pods:                         250
  rdma/rdma_shared_device_eth:  63
  rdma/rdma_shared_device_ib:   63
Allocatable:
  cpu:                          127500m
  ephemeral-storage:            1438028263499
  hugepages-1Gi:                0
  hugepages-2Mi:                0
  memory:                       262445736Ki
  nvidia.com/gpu:               2
  pods:                         250
  rdma/rdma_shared_device_eth:  63
  rdma/rdma_shared_device_ib:   63
--
Capacity:
  cpu:                          128
  ephemeral-storage:            1561525616Ki
  hugepages-1Gi:                0
  hugepages-2Mi:                0
  memory:                       263596672Ki
  nvidia.com/gpu:               2
  pods:                         250
  rdma/rdma_shared_device_eth:  63
  rdma/rdma_shared_device_ib:   63
Allocatable:
  cpu:                          127500m
  ephemeral-storage:            1438028263499
  hugepages-1Gi:                0
  hugepages-2Mi:                0
  memory:                       262445696Ki
  nvidia.com/gpu:               2
  pods:                         250
  rdma/rdma_shared_device_eth:  63
  rdma/rdma_shared_device_ib:   63

5.8. Creating the machine configuration
Copier lien

Before you create the resource pods, you need to create the machineconfig.yaml custom resource (CR) that provides access to the GPU and networking resources without the need for user privileges.

Procedure

Generate a Machineconfig CR:

apiVersion: machineconfiguration.openshift.io/v1
kind: MachineConfig
metadata:
  labels:
    machineconfiguration.openshift.io/role: worker
  name: 02-worker-container-runtime
spec:
  config:
    ignition:
      version: 3.2.0
    storage:
      files:
      - contents:
          source: data:text/plain;charset=utf-8;base64,W2NyaW8ucnVudGltZV0KZGVmYXVsdF91bGltaXRzID0gWwoibWVtbG9jaz0tMTotMSIKXQo=
        mode: 420
        overwrite: true
        path: /etc/crio/crio.conf.d/10-custom

5.9. Creating the workload pods
Copier lien

Use the procedures in this section to create the workload pods for the shared and host devices.

5.9.1. Creating a shared device RDMA on RoCE
Copier lien

Create the workload pods for a shared device RDMA on RDMA over Converged Ethernet (RoCE) for the NVIDIA Network Operator and test the pod configuration.

The NVIDIA GPUDirect RDMA device is shared among pods on the OpenShift Container Platform worker node where the device is exposed.

Prerequisites

Ensure that the Operator is running.
Delete the NicClusterPolicy custom resource (CR), if it exists.

Procedure

Generate custom pod resources:

$ cat <<EOF > rdma-eth-32-workload.yaml
apiVersion: v1
kind: Pod
metadata:
  name: rdma-eth-32-workload
  namespace: default
  annotations:
    k8s.v1.cni.cncf.io/networks: rdmashared-net
spec:
  nodeSelector:
    kubernetes.io/hostname: nvd-srv-32.nvidia.eng.rdu2.dc.redhat.com
  containers:
  - image: quay.io/edge-infrastructure/nvidia-tools:0.1.5
    name: rdma-eth-32-workload
    resources:
      limits:
        nvidia.com/gpu: 1
        rdma/rdma_shared_device_eth: 1
      requests:
        nvidia.com/gpu: 1
        rdma/rdma_shared_device_eth: 1

EOF

$ cat <<EOF > rdma-eth-33-workload.yaml
apiVersion: v1
kind: Pod
metadata:
  name: rdma-eth-33-workload
  namespace: default
  annotations:
    k8s.v1.cni.cncf.io/networks: rdmashared-net
spec:
  nodeSelector:
    kubernetes.io/hostname: nvd-srv-33.nvidia.eng.rdu2.dc.redhat.com
  containers:
  - image: quay.io/edge-infrastructure/nvidia-tools:0.1.5
    name: rdma-eth-33-workload
    securityContext:
      capabilities:
        add: [ "IPC_LOCK" ]
    resources:
      limits:
        nvidia.com/gpu: 1
        rdma/rdma_shared_device_eth: 1
      requests:
        nvidia.com/gpu: 1
        rdma/rdma_shared_device_eth: 1
EOF

Create the pods on the cluster by using the following commands:

$ oc create -f rdma-eth-32-workload.yaml

Example output

pod/rdma-eth-32-workload created

$ oc create -f rdma-eth-33-workload.yaml

Example output

pod/rdma-eth-33-workload created

Verify that the pods are running by using the following command:

$ oc get pods -n default

Example output

NAME                   READY   STATUS    RESTARTS   AGE
rdma-eth-32-workload   1/1     Running   0          25s
rdma-eth-33-workload   1/1     Running   0          22s

5.9.2. Creating a host device RDMA on RoCE
Copier lien

Create the workload pods for a host device Remote Direct Memory Access (RDMA) for the NVIDIA Network Operator and test the pod configuration.

Prerequisites

Ensure that the Operator is running.
Delete the NicClusterPolicy custom resource (CR), if it exists.

Procedure

Generate a new host device NicClusterPolicy (CR), as shown below:

$ cat <<EOF > network-hostdev-nic-cluster-policy.yaml
apiVersion: mellanox.com/v1alpha1
kind: NicClusterPolicy
metadata:
  name: nic-cluster-policy
spec:
  ofedDriver:
    image: doca-driver
    repository: nvcr.io/nvidia/mellanox
    version: 24.10-0.7.0.0-0
    startupProbe:
      initialDelaySeconds: 10
      periodSeconds: 20
    livenessProbe:
      initialDelaySeconds: 30
      periodSeconds: 30
    readinessProbe:
      initialDelaySeconds: 10
      periodSeconds: 30
    env:
    - name: UNLOAD_STORAGE_MODULES
      value: "true"
    - name: RESTORE_DRIVER_ON_POD_TERMINATION
      value: "true"
    - name: CREATE_IFNAMES_UDEV
      value: "true"
  sriovDevicePlugin:
      image: sriov-network-device-plugin
      repository: ghcr.io/k8snetworkplumbingwg
      version: v3.7.0
      config: |
        {
          "resourceList": [
              {
                  "resourcePrefix": "nvidia.com",
                  "resourceName": "hostdev",
                  "selectors": {
                      "vendors": ["15b3"],
                      "isRdma": true
                  }
              }
          ]
        }
EOF

Create the NicClusterPolicy CR on the cluster by using the following command:

$ oc create -f network-hostdev-nic-cluster-policy.yaml

Example output

nicclusterpolicy.mellanox.com/nic-cluster-policy created

Verify that the host device NicClusterPolicy CR by using the following command in the DOCA/MOFED container:

$ oc get pods -n nvidia-network-operator

Example output

NAME                                                          READY   STATUS    RESTARTS   AGE
mofed-rhcos4.16-696886fcb4-ds-9sgvd                           2/2     Running   0          2m37s
mofed-rhcos4.16-696886fcb4-ds-lkjd4                           2/2     Running   0          2m37s
nvidia-network-operator-controller-manager-68d547dbbd-qsdkf   1/1     Running   0          141m
sriov-device-plugin-6v2nz                                     1/1     Running   0          2m14s
sriov-device-plugin-hc4t8                                     1/1     Running   0          2m14s

Confirm that the resources appear in the cluster oc describe node section by using the following command:

$ oc describe node -l node-role.kubernetes.io/worker=| grep -E 'Capacity:|Allocatable:' -A7

Example output

Capacity:
  cpu:                 128
  ephemeral-storage:   1561525616Ki
  hugepages-1Gi:       0
  hugepages-2Mi:       0
  memory:              263596708Ki
  nvidia.com/hostdev:  2
  pods:                250
Allocatable:
  cpu:                 127500m
  ephemeral-storage:   1438028263499
  hugepages-1Gi:       0
  hugepages-2Mi:       0
  memory:              262445732Ki
  nvidia.com/hostdev:  2
  pods:                250
--
Capacity:
  cpu:                 128
  ephemeral-storage:   1561525616Ki
  hugepages-1Gi:       0
  hugepages-2Mi:       0
  memory:              263596704Ki
  nvidia.com/hostdev:  2
  pods:                250
Allocatable:
  cpu:                 127500m
  ephemeral-storage:   1438028263499
  hugepages-1Gi:       0
  hugepages-2Mi:       0
  memory:              262445728Ki
  nvidia.com/hostdev:  2
  pods:                250

Create a HostDeviceNetwork CR file:

$ cat <<EOF >  hostdev-network.yaml
apiVersion: mellanox.com/v1alpha1
kind: HostDeviceNetwork
metadata:
  name: hostdev-net
spec:
  networkNamespace: "default"
  resourceName: "hostdev"
  ipam: |
    {
      "type": "whereabouts",
      "range": "192.168.3.225/28",
      "exclude": [
       "192.168.3.229/30",
       "192.168.3.236/32"
      ]
    }
EOF

Create the HostDeviceNetwork resource on the cluster by using the following command:
```
$ oc create -f hostdev-network.yaml
```
Example output
```
hostdevicenetwork.mellanox.com/hostdev-net created
```

Confirm that the resources appear in the cluster oc describe node section by using the following command:

$ oc describe node -l node-role.kubernetes.io/worker=| grep -E 'Capacity:|Allocatable:' -A8

Example output

Capacity:
  cpu:                 128
  ephemeral-storage:   1561525616Ki
  hugepages-1Gi:       0
  hugepages-2Mi:       0
  memory:              263596708Ki
  nvidia.com/gpu:      2
  nvidia.com/hostdev:  2
  pods:                250
Allocatable:
  cpu:                 127500m
  ephemeral-storage:   1438028263499
  hugepages-1Gi:       0
  hugepages-2Mi:       0
  memory:              262445732Ki
  nvidia.com/gpu:      2
  nvidia.com/hostdev:  2
  pods:                250
--
Capacity:
  cpu:                 128
  ephemeral-storage:   1561525616Ki
  hugepages-1Gi:       0
  hugepages-2Mi:       0
  memory:              263596680Ki
  nvidia.com/gpu:      2
  nvidia.com/hostdev:  2
  pods:                250
Allocatable:
  cpu:                 127500m
  ephemeral-storage:   1438028263499
  hugepages-1Gi:       0
  hugepages-2Mi:       0
  memory:              262445704Ki
  nvidia.com/gpu:      2
  nvidia.com/hostdev:  2
  pods:                250

5.9.3. Creating a SR-IOV legacy mode RDMA on RoCE
Copier lien

Configure a Single Root I/O Virtualization (SR-IOV) legacy mode host device RDMA on RoCE.

Procedure

Generate a new host device NicClusterPolicy custom resource (CR):

$ cat <<EOF > network-sriovleg-nic-cluster-policy.yaml
apiVersion: mellanox.com/v1alpha1
kind: NicClusterPolicy
metadata:
  name: nic-cluster-policy
spec:
  ofedDriver:
    image: doca-driver
    repository: nvcr.io/nvidia/mellanox
    version: 24.10-0.7.0.0-0
    startupProbe:
      initialDelaySeconds: 10
      periodSeconds: 20
    livenessProbe:
      initialDelaySeconds: 30
      periodSeconds: 30
    readinessProbe:
      initialDelaySeconds: 10
      periodSeconds: 30
    env:
    - name: UNLOAD_STORAGE_MODULES
      value: "true"
    - name: RESTORE_DRIVER_ON_POD_TERMINATION
      value: "true"
    - name: CREATE_IFNAMES_UDEV
      value: "true"
EOF

Create the policy on the cluster by using the following command:

$ oc create -f network-sriovleg-nic-cluster-policy.yaml

Example output

nicclusterpolicy.mellanox.com/nic-cluster-policy created

Verify the pods by using the following command in the DOCA/MOFED container:

$ oc get pods -n nvidia-network-operator

Example output

NAME                                                          READY   STATUS    RESTARTS      AGE
mofed-rhcos4.16-696886fcb4-ds-4mb42                           2/2     Running   0             40s
mofed-rhcos4.16-696886fcb4-ds-8knwq                           2/2     Running   0             40s
nvidia-network-operator-controller-manager-68d547dbbd-qsdkf   1/1     Running   13 (4d ago)   4d21h

Create an SriovNetworkNodePolicy CR that generates the Virtual Functions (VFs) for the device you want to operate in SR-IOV legacy mode. See the following example:

$ cat <<EOF > sriov-network-node-policy.yaml
apiVersion: sriovnetwork.openshift.io/v1
kind: SriovNetworkNodePolicy
metadata:
  name: sriov-legacy-policy
  namespace:  openshift-sriov-network-operator
spec:
  deviceType: netdevice
  mtu: 1500
  nicSelector:
    vendor: "15b3"
    pfNames: ["ens8f0np0#0-7"]
  nodeSelector:
    feature.node.kubernetes.io/pci-15b3.present: "true"
  numVfs: 8
  priority: 90
  isRdma: true
  resourceName: sriovlegacy
EOF

Create the CR on the cluster by using the following command:
Note
Ensure that SR-IOV Global Enable is enabled. For more information, see Unable to enable SR-IOV and receiving the message "not enough MMIO resources for SR-IOV" in Red Hat Enterprise Linux.
```
$ oc create -f sriov-network-node-policy.yaml
```
Example output
```
sriovnetworknodepolicy.sriovnetwork.openshift.io/sriov-legacy-policy created
```

Each node has scheduling disabled. The nodes reboot to apply the configuration. You can view the nodes by using the following command:

$ oc get nodes

Example output

NAME                                       STATUS                        ROLES                         AGE     VERSION
edge-19.edge.lab.eng.rdu2.redhat.com       Ready                         control-plane,master,worker   5d      v1.29.8+632b078
nvd-srv-32.nvidia.eng.rdu2.dc.redhat.com   Ready                         worker                        4d22h   v1.29.8+632b078
nvd-srv-33.nvidia.eng.rdu2.dc.redhat.com   NotReady,SchedulingDisabled   worker                        4d22h   v1.29.8+632b078

After the nodes have rebooted, verify that the VF interfaces exist by opening up a debug pod on each node. Run the following command:

a$ oc debug node/nvd-srv-33.nvidia.eng.rdu2.dc.redhat.com

Example output

Starting pod/nvd-srv-33nvidiaengrdu2dcredhatcom-debug-cqfjz ...
To use host binaries, run `chroot /host`
Pod IP: 10.6.135.12
If you don't see a command prompt, try pressing enter.
sh-5.1# chroot /host
sh-5.1# ip link show | grep ens8
26: ens8f0np0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000
42: ens8f0v0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
43: ens8f0v1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
44: ens8f0v2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
45: ens8f0v3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
46: ens8f0v4: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
47: ens8f0v5: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
48: ens8f0v6: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
49: ens8f0v7: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000

Repeat the previous steps on the second node, if necessary.

Optional: Confirm that the resources appear in the cluster oc describe node section by using the following command:

$ oc describe node -l node-role.kubernetes.io/worker=| grep -E 'Capacity:|Allocatable:' -A8

Example output

Capacity:
  cpu:                       128
  ephemeral-storage:         1561525616Ki
  hugepages-1Gi:             0
  hugepages-2Mi:             0
  memory:                    263596692Ki
  nvidia.com/gpu:            2
  nvidia.com/hostdev:        0
  openshift.io/sriovlegacy:  8
--
Allocatable:
  cpu:                       127500m
  ephemeral-storage:         1438028263499
  hugepages-1Gi:             0
  hugepages-2Mi:             0
  memory:                    262445716Ki
  nvidia.com/gpu:            2
  nvidia.com/hostdev:        0
  openshift.io/sriovlegacy:  8
--
Capacity:
  cpu:                       128
  ephemeral-storage:         1561525616Ki
  hugepages-1Gi:             0
  hugepages-2Mi:             0
  memory:                    263596688Ki
  nvidia.com/gpu:            2
  nvidia.com/hostdev:        0
  openshift.io/sriovlegacy:  8
--
Allocatable:
  cpu:                       127500m
  ephemeral-storage:         1438028263499
  hugepages-1Gi:             0
  hugepages-2Mi:             0
  memory:                    262445712Ki
  nvidia.com/gpu:            2
  nvidia.com/hostdev:        0
  openshift.io/sriovlegacy:  8

After the VFs for SR-IOV legacy mode are in place, generate the SriovNetwork CR file. See the following example:

$ cat <<EOF > sriov-network.yaml
apiVersion: sriovnetwork.openshift.io/v1
kind: SriovNetwork
metadata:
  name: sriov-network
  namespace:  openshift-sriov-network-operator
spec:
  vlan: 0
  networkNamespace: "default"
  resourceName: "sriovlegacy"
  ipam: |
    {
      "type": "whereabouts",
      "range": "192.168.3.225/28",
      "exclude": [
       "192.168.3.229/30",
       "192.168.3.236/32"
      ]
    }
EOF

Create the custom resource on the cluster by using the following command:

$ oc create -f sriov-network.yaml

Example output

sriovnetwork.sriovnetwork.openshift.io/sriov-network created

5.9.4. Creating a shared device RDMA on Infiniband
Copier lien

Create the workload pods for a shared device Remote Direct Memory Access (RDMA) for an Infiniband installation.

Procedure

Generate custom pod resources:

$ cat <<EOF > rdma-ib-32-workload.yaml
apiVersion: v1
kind: Pod
metadata:
  name: rdma-ib-32-workload
  namespace: default
  annotations:
    k8s.v1.cni.cncf.io/networks: example-ipoibnetwork
spec:
  nodeSelector:
    kubernetes.io/hostname: nvd-srv-32.nvidia.eng.rdu2.dc.redhat.com
  containers:
  - image: quay.io/edge-infrastructure/nvidia-tools:0.1.5
    name: rdma-ib-32-workload
    resources:
      limits:
        nvidia.com/gpu: 1
        rdma/rdma_shared_device_ib: 1
      requests:
        nvidia.com/gpu: 1
        rdma/rdma_shared_device_ib: 1
EOF

$ cat <<EOF > rdma-ib-32-workload.yaml
apiVersion: v1
kind: Pod
metadata:
  name: rdma-ib-33-workload
  namespace: default
  annotations:
    k8s.v1.cni.cncf.io/networks: example-ipoibnetwork
spec:
  nodeSelector:
    kubernetes.io/hostname: nvd-srv-33.nvidia.eng.rdu2.dc.redhat.com
  containers:
  - image: quay.io/edge-infrastructure/nvidia-tools:0.1.5
    name: rdma-ib-33-workload
    securityContext:
      capabilities:
        add: [ "IPC_LOCK" ]
    resources:
      limits:
        nvidia.com/gpu: 1
        rdma/rdma_shared_device_ib: 1
      requests:
        nvidia.com/gpu: 1
        rdma/rdma_shared_device_ib: 1
EOF

Create the pods on the cluster by using the following commands:

$ oc create -f rdma-ib-32-workload.yaml

Example output

pod/rdma-ib-32-workload created

$ oc create -f rdma-ib-33-workload.yaml

Example output

pod/rdma-ib-33-workload created

Verify that the pods are running by using the following command:

$ oc get pods

Example output

NAME                  READY   STATUS    RESTARTS   AGE
rdma-ib-32-workload   1/1     Running   0          10s
rdma-ib-33-workload   1/1     Running   0          3s

5.10. Verifying RDMA connectivity
Copier lien

Confirm Remote Direct Memory Access (RDMA) connectivity is working between the systems, specifically for Legacy Single Root I/O Virtualization (SR-IOV) Ethernet.

Procedure

Connect to each rdma-workload-client pod by using the following command:
```
$ oc rsh -n default rdma-sriov-32-workload
```
Example output
```
sh-5.1#
```

Check the IP address assigned to the first workload pod by using the following command. In this example, the first workload pod is the RDMA test server.

sh-5.1# ip a

Example output

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0@if3970: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1400 qdisc noqueue state UP group default
    link/ether 0a:58:0a:80:02:a7 brd ff:ff:ff:ff:ff:ff link-netnsid 0
    inet 10.128.2.167/23 brd 10.128.3.255 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::858:aff:fe80:2a7/64 scope link
       valid_lft forever preferred_lft forever
3843: net1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 26:34:fd:53:a6:ec brd ff:ff:ff:ff:ff:ff
    altname enp55s0f0v5
    inet 192.168.4.225/28 brd 192.168.4.239 scope global net1
       valid_lft forever preferred_lft forever
    inet6 fe80::2434:fdff:fe53:a6ec/64 scope link
       valid_lft forever preferred_lft forever
sh-5.1#

The IP address of the RDMA server assigned to this pod is the net1 interface. In this example, the IP address is 192.168.4.225.

Run the ibstatus command to get the link_layer type, Ethernet or Infiniband, associated with each RDMA device mlx5_x. The output also shows the status of all of the RDMA devices by checking the state field, which shows either ACTIVE or DOWN.

sh-5.1# ibstatus

Example output

Infiniband device 'mlx5_0' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 4: ACTIVE
	phys state:	 5: LinkUp
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_1' port 1 status:
	default gid:	 fe80:0000:0000:0000:e8eb:d303:0072:1415
	base lid:	 0xc
	sm lid:		 0x1
	state:		 4: ACTIVE
	phys state:	 5: LinkUp
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 InfiniBand

Infiniband device 'mlx5_2' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_3' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_4' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_5' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_6' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_7' port 1 status:
	default gid:	 fe80:0000:0000:0000:2434:fdff:fe53:a6ec
	base lid:	 0x0
	sm lid:		 0x0
	state:		 4: ACTIVE
	phys state:	 5: LinkUp
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_8' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_9' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

sh-5.1#

To get the link_layer for each RDMA mlx5 device on your worker node, run the ibstat command:

sh-5.1# ibstat | egrep "Port|Base|Link"

Example output

Port 1:
		Physical state: LinkUp
		Base lid: 0
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
	Port 1:
		Physical state: LinkUp
		Base lid: 12
		Port GUID: 0xe8ebd30300721415
		Link layer: InfiniBand
	Port 1:
		Base lid: 0
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
	Port 1:
		Base lid: 0
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
	Port 1:
		Base lid: 0
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
	Port 1:
		Base lid: 0
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
	Port 1:
		Base lid: 0
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
	Port 1:
		Physical state: LinkUp
		Base lid: 0
		Port GUID: 0x2434fdfffe53a6ec
		Link layer: Ethernet
	Port 1:
		Base lid: 0
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
	Port 1:
		Base lid: 0
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
sh-5.1#

For RDMA Shared Device or Host Device workload pods, the RDMA device named mlx5_x is already known and is typically mlx5_0 or mlx5_1. For RDMA Legacy SR-IOV workload pods, you need to determine which RDMA device is associated with which Virtual Function (VF) subinterface. Provide this information by using the following command:

sh-5.1# rdma link show

Example output

link mlx5_0/1 state ACTIVE physical_state LINK_UP
link mlx5_1/1 subnet_prefix fe80:0000:0000:0000 lid 12 sm_lid 1 lmc 0 state ACTIVE physical_state LINK_UP
link mlx5_2/1 state DOWN physical_state DISABLED
link mlx5_3/1 state DOWN physical_state DISABLED
link mlx5_4/1 state DOWN physical_state DISABLED
link mlx5_5/1 state DOWN physical_state DISABLED
link mlx5_6/1 state DOWN physical_state DISABLED
link mlx5_7/1 state ACTIVE physical_state LINK_UP netdev net1
link mlx5_8/1 state DOWN physical_state DISABLED
link mlx5_9/1 state DOWN physical_state DISABLED

In this example, the RDMA device names mlx5_7 is associated with the net1 interface. This output is used in the next command to perform the RDMA bandwidth test, which also verifies RDMA connectivity between worker nodes.

Run the following ib_write_bw RDMA bandwidth test command:

sh-5.1# /root/perftest/ib_write_bw -R -T 41 -s 65536 -F -x 3 -m 4096 --report_gbits -q 16 -D 60  -d mlx5_7 -p 10000 --source_ip  192.168.4.225 --use_cuda=0 --use_cuda_dmabuf

where:

The mlx5_7 RDMA device is passed in the -d switch.
The source IP address is 192.168.4.225 to start the RDMA server.
The --use_cuda=0, --use_cuda_dmabuf switches indicate that the use of GPUDirect RDMA.

Example output

WARNING: BW peak won't be measured in this run.
Perftest doesn't supports CUDA tests with inline messages: inline size set to 0

************************************
* Waiting for client to connect... *
************************************

Open another terminal window and run oc rsh command on the second workload pod that acts as the RDMA test client pod:
```
$ oc rsh -n default rdma-sriov-33-workload
```
Example output
```
sh-5.1#
```

Obtain the RDMA test client pod IP address from the net1 interface by using the following command:

sh-5.1# ip a

Example output

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0@if4139: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1400 qdisc noqueue state UP group default
    link/ether 0a:58:0a:83:01:d5 brd ff:ff:ff:ff:ff:ff link-netnsid 0
    inet 10.131.1.213/23 brd 10.131.1.255 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::858:aff:fe83:1d5/64 scope link
       valid_lft forever preferred_lft forever
4076: net1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 56:6c:59:41:ae:4a brd ff:ff:ff:ff:ff:ff
    altname enp55s0f0v0
    inet 192.168.4.226/28 brd 192.168.4.239 scope global net1
       valid_lft forever preferred_lft forever
    inet6 fe80::546c:59ff:fe41:ae4a/64 scope link
       valid_lft forever preferred_lft forever
sh-5.1#

Obtain the link_layer type associated with each RDMA device mlx5_x by using the following command:

sh-5.1# ibstatus

Example output

Infiniband device 'mlx5_0' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 4: ACTIVE
	phys state:	 5: LinkUp
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_1' port 1 status:
	default gid:	 fe80:0000:0000:0000:e8eb:d303:0072:09f5
	base lid:	 0xd
	sm lid:		 0x1
	state:		 4: ACTIVE
	phys state:	 5: LinkUp
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 InfiniBand

Infiniband device 'mlx5_2' port 1 status:
	default gid:	 fe80:0000:0000:0000:546c:59ff:fe41:ae4a
	base lid:	 0x0
	sm lid:		 0x0
	state:		 4: ACTIVE
	phys state:	 5: LinkUp
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_3' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_4' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_5' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_6' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_7' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_8' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Infiniband device 'mlx5_9' port 1 status:
	default gid:	 0000:0000:0000:0000:0000:0000:0000:0000
	base lid:	 0x0
	sm lid:		 0x0
	state:		 1: DOWN
	phys state:	 3: Disabled
	rate:		 200 Gb/sec (4X HDR)
	link_layer:	 Ethernet

Optional: Obtain the firmware version of Mellanox cards by using the ibstat command:

sh-5.1# ibstat

Example output

CA 'mlx5_0'
	CA type: MT4123
	Number of ports: 1
	Firmware version: 20.43.1014
	Hardware version: 0
	Node GUID: 0xe8ebd303007209f4
	System image GUID: 0xe8ebd303007209f4
	Port 1:
		State: Active
		Physical state: LinkUp
		Rate: 200
		Base lid: 0
		LMC: 0
		SM lid: 0
		Capability mask: 0x00010000
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
CA 'mlx5_1'
	CA type: MT4123
	Number of ports: 1
	Firmware version: 20.43.1014
	Hardware version: 0
	Node GUID: 0xe8ebd303007209f5
	System image GUID: 0xe8ebd303007209f4
	Port 1:
		State: Active
		Physical state: LinkUp
		Rate: 200
		Base lid: 13
		LMC: 0
		SM lid: 1
		Capability mask: 0xa651e848
		Port GUID: 0xe8ebd303007209f5
		Link layer: InfiniBand
CA 'mlx5_2'
	CA type: MT4124
	Number of ports: 1
	Firmware version: 20.43.1014
	Hardware version: 0
	Node GUID: 0x566c59fffe41ae4a
	System image GUID: 0xe8ebd303007209f4
	Port 1:
		State: Active
		Physical state: LinkUp
		Rate: 200
		Base lid: 0
		LMC: 0
		SM lid: 0
		Capability mask: 0x00010000
		Port GUID: 0x546c59fffe41ae4a
		Link layer: Ethernet
CA 'mlx5_3'
	CA type: MT4124
	Number of ports: 1
	Firmware version: 20.43.1014
	Hardware version: 0
	Node GUID: 0xb2ae4bfffe8f3d02
	System image GUID: 0xe8ebd303007209f4
	Port 1:
		State: Down
		Physical state: Disabled
		Rate: 200
		Base lid: 0
		LMC: 0
		SM lid: 0
		Capability mask: 0x00010000
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
CA 'mlx5_4'
	CA type: MT4124
	Number of ports: 1
	Firmware version: 20.43.1014
	Hardware version: 0
	Node GUID: 0x2a9967fffe8bf272
	System image GUID: 0xe8ebd303007209f4
	Port 1:
		State: Down
		Physical state: Disabled
		Rate: 200
		Base lid: 0
		LMC: 0
		SM lid: 0
		Capability mask: 0x00010000
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
CA 'mlx5_5'
	CA type: MT4124
	Number of ports: 1
	Firmware version: 20.43.1014
	Hardware version: 0
	Node GUID: 0x5aff2ffffe2e17e8
	System image GUID: 0xe8ebd303007209f4
	Port 1:
		State: Down
		Physical state: Disabled
		Rate: 200
		Base lid: 0
		LMC: 0
		SM lid: 0
		Capability mask: 0x00010000
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
CA 'mlx5_6'
	CA type: MT4124
	Number of ports: 1
	Firmware version: 20.43.1014
	Hardware version: 0
	Node GUID: 0x121bf1fffe074419
	System image GUID: 0xe8ebd303007209f4
	Port 1:
		State: Down
		Physical state: Disabled
		Rate: 200
		Base lid: 0
		LMC: 0
		SM lid: 0
		Capability mask: 0x00010000
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
CA 'mlx5_7'
	CA type: MT4124
	Number of ports: 1
	Firmware version: 20.43.1014
	Hardware version: 0
	Node GUID: 0xb22b16fffed03dd7
	System image GUID: 0xe8ebd303007209f4
	Port 1:
		State: Down
		Physical state: Disabled
		Rate: 200
		Base lid: 0
		LMC: 0
		SM lid: 0
		Capability mask: 0x00010000
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
CA 'mlx5_8'
	CA type: MT4124
	Number of ports: 1
	Firmware version: 20.43.1014
	Hardware version: 0
	Node GUID: 0x523800fffe16d105
	System image GUID: 0xe8ebd303007209f4
	Port 1:
		State: Down
		Physical state: Disabled
		Rate: 200
		Base lid: 0
		LMC: 0
		SM lid: 0
		Capability mask: 0x00010000
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
CA 'mlx5_9'
	CA type: MT4124
	Number of ports: 1
	Firmware version: 20.43.1014
	Hardware version: 0
	Node GUID: 0xd2b4a1fffebdc4a9
	System image GUID: 0xe8ebd303007209f4
	Port 1:
		State: Down
		Physical state: Disabled
		Rate: 200
		Base lid: 0
		LMC: 0
		SM lid: 0
		Capability mask: 0x00010000
		Port GUID: 0x0000000000000000
		Link layer: Ethernet
sh-5.1#

To determine which RDMA device is associated with the Virtual Function subinterface that the client workload pod uses, run the following command. In this example, the net1 interface is using the RDMA device mlx5_2.

sh-5.1# rdma link show

Example output

link mlx5_0/1 state ACTIVE physical_state LINK_UP
link mlx5_1/1 subnet_prefix fe80:0000:0000:0000 lid 13 sm_lid 1 lmc 0 state ACTIVE physical_state LINK_UP
link mlx5_2/1 state ACTIVE physical_state LINK_UP netdev net1
link mlx5_3/1 state DOWN physical_state DISABLED
link mlx5_4/1 state DOWN physical_state DISABLED
link mlx5_5/1 state DOWN physical_state DISABLED
link mlx5_6/1 state DOWN physical_state DISABLED
link mlx5_7/1 state DOWN physical_state DISABLED
link mlx5_8/1 state DOWN physical_state DISABLED
link mlx5_9/1 state DOWN physical_state DISABLED
sh-5.1#

Run the following ib_write_bw RDMA bandwidth test command:

sh-5.1# /root/perftest/ib_write_bw -R -T 41 -s 65536 -F -x 3 -m 4096 --report_gbits -q 16 -D 60  -d mlx5_2 -p 10000 --source_ip  192.168.4.226 --use_cuda=0 --use_cuda_dmabuf 192.168.4.225

where:

The mlx5_2 RDMA device is passed in the -d switch.
The source IP address 192.168.4.226 and the destination IP address of the RDMA server 192.168.4.225.

The --use_cuda=0, --use_cuda_dmabuf switches indicate that the use of GPUDirect RDMA.

Example output

WARNING: BW peak won't be measured in this run.
Perftest doesn't supports CUDA tests with inline messages: inline size set to 0
Requested mtu is higher than active mtu
Changing to active mtu - 3
initializing CUDA
Listing all CUDA devices in system:
CUDA device 0: PCIe address is 61:00

Picking device No. 0
[pid = 8909, dev = 0] device name = [NVIDIA A40]
creating CUDA Ctx
making it the current CUDA Ctx
CUDA device integrated: 0
using DMA-BUF for GPU buffer address at 0x7f8738600000 aligned at 0x7f8738600000 with aligned size 2097152
allocated GPU buffer of a 2097152 address at 0x23a7420 for type CUDA_MEM_DEVICE
Calling ibv_reg_dmabuf_mr(offset=0, size=2097152, addr=0x7f8738600000, fd=40) for QP #0
---------------------------------------------------------------------------------------
                    RDMA_Write BW Test
 Dual-port       : OFF		Device         : mlx5_2
 Number of qps   : 16		Transport type : IB
 Connection type : RC		Using SRQ      : OFF
 PCIe relax order: ON		Lock-free      : OFF
 ibv_wr* API     : ON		Using DDP      : OFF
 TX depth        : 128
 CQ Moderation   : 1
 CQE Poll Batch  : 16
 Mtu             : 1024[B]
 Link type       : Ethernet
 GID index       : 3
 Max inline data : 0[B]
 rdma_cm QPs	 : ON
 Data ex. method : rdma_cm 	TOS    : 41
---------------------------------------------------------------------------------------
 local address: LID 0000 QPN 0x012d PSN 0x3cb6d7
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x012e PSN 0x90e0ac
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x012f PSN 0x153f50
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x0130 PSN 0x5e0128
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x0131 PSN 0xd89752
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x0132 PSN 0xe5fc16
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x0133 PSN 0x236787
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x0134 PSN 0xd9273e
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x0135 PSN 0x37cfd4
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x0136 PSN 0x3bff8f
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x0137 PSN 0x81f2bd
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x0138 PSN 0x575c43
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x0139 PSN 0x6cf53d
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x013a PSN 0xcaaf6f
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x013b PSN 0x346437
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 local address: LID 0000 QPN 0x013c PSN 0xcc5865
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x026d PSN 0x359409
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x026e PSN 0xe387bf
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x026f PSN 0x5be79d
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x0270 PSN 0x1b4b28
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x0271 PSN 0x76a61b
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x0272 PSN 0x3d50e1
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x0273 PSN 0x1b572c
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x0274 PSN 0x4ae1b5
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x0275 PSN 0x5591b5
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x0276 PSN 0xfa2593
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x0277 PSN 0xd9473b
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x0278 PSN 0x2116b2
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x0279 PSN 0x9b83b6
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x027a PSN 0xa0822b
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x027b PSN 0x6d930d
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x027c PSN 0xb1a4d
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
---------------------------------------------------------------------------------------
 #bytes     #iterations    BW peak[Gb/sec]    BW average[Gb/sec]   MsgRate[Mpps]
 65536      10329004         0.00               180.47 		     0.344228
---------------------------------------------------------------------------------------
deallocating GPU buffer 00007f8738600000
destroying current CUDA Ctx
sh-5.1#

A positive test is seeing an expected BW average and MsgRate in Mpps.

Upon completion of the ib_write_bw command, the server side output also appears on the server pod. See the following example:

Example output

WARNING: BW peak won't be measured in this run.
Perftest doesn't supports CUDA tests with inline messages: inline size set to 0

************************************
* Waiting for client to connect... *
************************************
Requested mtu is higher than active mtu
Changing to active mtu - 3
initializing CUDA
Listing all CUDA devices in system:
CUDA device 0: PCIe address is 61:00

Picking device No. 0
[pid = 9226, dev = 0] device name = [NVIDIA A40]
creating CUDA Ctx
making it the current CUDA Ctx
CUDA device integrated: 0
using DMA-BUF for GPU buffer address at 0x7f447a600000 aligned at 0x7f447a600000 with aligned size 2097152
allocated GPU buffer of a 2097152 address at 0x2406400 for type CUDA_MEM_DEVICE
Calling ibv_reg_dmabuf_mr(offset=0, size=2097152, addr=0x7f447a600000, fd=40) for QP #0
---------------------------------------------------------------------------------------
                    RDMA_Write BW Test
 Dual-port       : OFF		Device         : mlx5_7
 Number of qps   : 16		Transport type : IB
 Connection type : RC		Using SRQ      : OFF
 PCIe relax order: ON		Lock-free      : OFF
 ibv_wr* API     : ON		Using DDP      : OFF
 CQ Moderation   : 1
 CQE Poll Batch  : 16
 Mtu             : 1024[B]
 Link type       : Ethernet
 GID index       : 3
 Max inline data : 0[B]
 rdma_cm QPs	 : ON
 Data ex. method : rdma_cm 	TOS    : 41
---------------------------------------------------------------------------------------
 Waiting for client rdma_cm QP to connect
 Please run the same command with the IB/RoCE interface IP
---------------------------------------------------------------------------------------
 local address: LID 0000 QPN 0x026d PSN 0x359409
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x026e PSN 0xe387bf
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x026f PSN 0x5be79d
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x0270 PSN 0x1b4b28
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x0271 PSN 0x76a61b
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x0272 PSN 0x3d50e1
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x0273 PSN 0x1b572c
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x0274 PSN 0x4ae1b5
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x0275 PSN 0x5591b5
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x0276 PSN 0xfa2593
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x0277 PSN 0xd9473b
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x0278 PSN 0x2116b2
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x0279 PSN 0x9b83b6
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x027a PSN 0xa0822b
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x027b PSN 0x6d930d
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 local address: LID 0000 QPN 0x027c PSN 0xb1a4d
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:225
 remote address: LID 0000 QPN 0x012d PSN 0x3cb6d7
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x012e PSN 0x90e0ac
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x012f PSN 0x153f50
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x0130 PSN 0x5e0128
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x0131 PSN 0xd89752
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x0132 PSN 0xe5fc16
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x0133 PSN 0x236787
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x0134 PSN 0xd9273e
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x0135 PSN 0x37cfd4
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x0136 PSN 0x3bff8f
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x0137 PSN 0x81f2bd
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x0138 PSN 0x575c43
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x0139 PSN 0x6cf53d
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x013a PSN 0xcaaf6f
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x013b PSN 0x346437
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
 remote address: LID 0000 QPN 0x013c PSN 0xcc5865
 GID: 00:00:00:00:00:00:00:00:00:00:255:255:192:168:04:226
---------------------------------------------------------------------------------------
 #bytes     #iterations    BW peak[Gb/sec]    BW average[Gb/sec]   MsgRate[Mpps]
 65536      10329004         0.00               180.47 		     0.344228
---------------------------------------------------------------------------------------
deallocating GPU buffer 00007f447a600000
destroying current CUDA Ctx

Ce contenu n'est pas disponible dans la langue sélectionnée.

Chapter 5. NVIDIA GPUDirect Remote Direct Memory Access (RDMA)

5.1. NVIDIA GPUDirect RDMA prerequisites
Copier lien

5.2. Disabling the IRDMA kernel module
Copier lien

5.3. Creating persistent naming rules
Copier lien

5.4. Configuring the NFD Operator
Copier lien

5.5. Configuring the SR-IOV Operator
Copier lien

5.6. Configuring the NVIDIA network Operator
Copier lien

5.7. Configuring the GPU Operator
Copier lien

5.8. Creating the machine configuration
Copier lien

5.9. Creating the workload pods
Copier lien

5.9.1. Creating a shared device RDMA on RoCE
Copier lien

5.9.2. Creating a host device RDMA on RoCE
Copier lien

5.9.3. Creating a SR-IOV legacy mode RDMA on RoCE
Copier lien

5.9.4. Creating a shared device RDMA on Infiniband
Copier lien

5.10. Verifying RDMA connectivity
Copier lien

Apprendre

Essayez, achetez et vendez

Communautés

À propos de la documentation Red Hat

Rendre l’open source plus inclusif

À propos de Red Hat

Theme

Red Hat legal and privacy links

Red Hat legal and privacy links

Ce contenu n'est pas disponible dans la langue sélectionnée.

Chapter 5. NVIDIA GPUDirect Remote Direct Memory Access (RDMA)

5.1. NVIDIA GPUDirect RDMA prerequisitesCopier lienLien copié sur presse-papiers!

5.2. Disabling the IRDMA kernel moduleCopier lienLien copié sur presse-papiers!

5.3. Creating persistent naming rulesCopier lienLien copié sur presse-papiers!

5.4. Configuring the NFD OperatorCopier lienLien copié sur presse-papiers!

5.5. Configuring the SR-IOV OperatorCopier lienLien copié sur presse-papiers!

5.6. Configuring the NVIDIA network OperatorCopier lienLien copié sur presse-papiers!

5.7. Configuring the GPU OperatorCopier lienLien copié sur presse-papiers!

5.8. Creating the machine configurationCopier lienLien copié sur presse-papiers!

5.9. Creating the workload podsCopier lienLien copié sur presse-papiers!

5.9.1. Creating a shared device RDMA on RoCECopier lienLien copié sur presse-papiers!

5.9.2. Creating a host device RDMA on RoCECopier lienLien copié sur presse-papiers!

5.9.3. Creating a SR-IOV legacy mode RDMA on RoCECopier lienLien copié sur presse-papiers!

5.9.4. Creating a shared device RDMA on InfinibandCopier lienLien copié sur presse-papiers!

5.10. Verifying RDMA connectivityCopier lienLien copié sur presse-papiers!

Apprendre

Essayez, achetez et vendez

Communautés

À propos de la documentation Red Hat

Rendre l’open source plus inclusif

À propos de Red Hat

Theme

Red Hat legal and privacy links

Red Hat legal and privacy links

5.1. NVIDIA GPUDirect RDMA prerequisites
Copier lien

5.2. Disabling the IRDMA kernel module
Copier lien

5.3. Creating persistent naming rules
Copier lien

5.4. Configuring the NFD Operator
Copier lien

5.5. Configuring the SR-IOV Operator
Copier lien

5.6. Configuring the NVIDIA network Operator
Copier lien

5.7. Configuring the GPU Operator
Copier lien

5.8. Creating the machine configuration
Copier lien

5.9. Creating the workload pods
Copier lien

5.9.1. Creating a shared device RDMA on RoCE
Copier lien

5.9.2. Creating a host device RDMA on RoCE
Copier lien

5.9.3. Creating a SR-IOV legacy mode RDMA on RoCE
Copier lien

5.9.4. Creating a shared device RDMA on Infiniband
Copier lien

5.10. Verifying RDMA connectivity
Copier lien