Create a distributed training.

POST

http://localhost:8080/api/v1/workloads/distributed

POST

/api/v1/workloads/distributed

curl -X POST http://localhost:8080/api/v1/workloads/distributed \
     -H "Authorization: Bearer <token>" \
     -H "Content-Type: application/json" \
     -d '{
  "name": "my-workload-name",
  "projectId": "1",
  "clusterId": "71f69d83-ba66-4822-adf5-55ce55efd210"
}'

{
  "name": "my-workload-name",
  "requestedName": "string",
  "workloadId": "string",
  "projectId": "1",
  "clusterId": "71f69d83-ba66-4822-adf5-55ce55efd210",
  "createdBy": "test@lab.com",
  "createdAt": "2022-01-01T03:49:52.531Z",
  "desiredPhase": "Running",
  "departmentId": "2",
  "deletedAt": "2022-01-01T03:49:52.531Z",
  "actualPhase": "Creating",
  "spec": {
    "annotations": [
      {
        "name": "billing",
        "value": "my-billing-unit",
        "exclude": false
      }
    ],
    "args": "-x my-script.py",
    "autoDeletionTimeAfterCompletionSeconds": 15,
    "backoffLimit": 3,
    "category": "string",
    "cleanPodPolicy": "None",
    "command": "python",
    "compute": {
      "cpuCoreLimit": 2,
      "cpuCoreRequest": 0.5,
      "cpuMemoryLimit": "30M",
      "cpuMemoryRequest": "20M",
      "extendedResources": [
        {
          "resource": "hardware-vendor.example/foo",
          "quantity": "2",
          "exclude": false
        }
      ],
      "gpuDevicesRequest": 1,
      "gpuMemoryLimit": "10M",
      "gpuMemoryRequest": "10M",
      "gpuPortionLimit": 0.5,
      "gpuPortionRequest": 0.5,
      "gpuRequestType": "portion",
      "largeShmRequest": false
    },
    "createHomeDir": true,
    "distributedFramework": "MPI",
    "environmentVariables": [
      {
        "name": "HOME",
        "value": "/home/my-folder",
        "secret": {
          "name": "postgress_secret",
          "key": "POSTGRES_PASSWORD"
        },
        "configMap": {
          "name": "my-config-map",
          "key": "MY_POSTGRES_SCHEMA"
        },
        "podFieldRef": {
          "path": "metadata.name"
        },
        "userCredential": {
          "name": "my_postgres_user_and_password",
          "key": "POSTGRES_PASSWORD"
        },
        "exclude": false,
        "description": "Home directory of the user."
      }
    ],
    "exposedUrls": [
      {
        "container": 8080,
        "url": "https://my-url.com",
        "authorizationType": "authenticatedUsers",
        "authorizedUsers": [
          "user-a",
          "user-b"
        ],
        "authorizedGroups": [
          "group-a",
          "group-b"
        ],
        "toolType": "jupyter",
        "toolName": "my-pytorch",
        "name": "url-instance-a",
        "exclude": false
      }
    ],
    "image": "python:3.8",
    "imagePullPolicy": "Always",
    "imagePullSecrets": [
      {
        "name": "string",
        "userCredential": true,
        "exclude": false
      }
    ],
    "labels": [
      {
        "name": "stage",
        "value": "initial-research",
        "exclude": false
      }
    ],
    "maxReplicas": 1,
    "minReplicas": 1,
    "mpiLauncherCreationPolicy": "AtStartup",
    "nodeAffinityRequired": {
      "nodeSelectorTerms": [
        {
          "matchExpressions": [
            {
              "key": "string",
              "operator": "In",
              "values": [
                "string"
              ]
            }
          ]
        }
      ]
    },
    "nodePools": [
      "my-node-pool-a",
      "my-node-pool-b"
    ],
    "nodeType": "my-node-type",
    "numWorkers": 1,
    "podAffinity": {
      "type": "Required",
      "key": "string"
    },
    "ports": [
      {
        "container": 8080,
        "serviceType": "LoadBalancer",
        "external": 30080,
        "toolType": "pytorch",
        "toolName": "my-pytorch",
        "name": "port-instance-a",
        "exclude": false
      }
    ],
    "preemptibility": "preemptible",
    "priorityClass": "string",
    "probes": {
      "readiness": {
        "initialDelaySeconds": 1,
        "periodSeconds": 1,
        "timeoutSeconds": 1,
        "successThreshold": 1,
        "failureThreshold": 1,
        "handler": {
          "httpGet": {
            "path": "/",
            "port": 1,
            "host": "example.com",
            "scheme": "HTTP"
          }
        }
      }
    },
    "relatedUrls": [
      {
        "url": "https://my-url.com",
        "type": "wandb",
        "name": "url-instance-a",
        "exclude": false
      }
    ],
    "restartPolicy": "Always",
    "security": {
      "allowPrivilegeEscalation": false,
      "capabilities": [
        "CHOWN",
        "KILL"
      ],
      "hostIpc": false,
      "hostNetwork": false,
      "readOnlyRootFilesystem": false,
      "runAsGid": 30,
      "runAsNonRoot": true,
      "runAsUid": 500,
      "seccompProfileType": "RuntimeDefault",
      "supplementalGroups": "2,3,5,8",
      "uidGidSource": "fromTheImage"
    },
    "segmentSize": 1,
    "slotsPerWorker": 1,
    "sshAuthMountPath": "/root/.ssh",
    "stdin": true,
    "storage": {
      "configMapVolume": [
        {
          "configMap": "string",
          "defaultMode": "0644",
          "exclude": false,
          "mountPath": "string",
          "name": "storage-instance-a",
          "subPath": "string"
        }
      ],
      "dataVolume": [
        {
          "exclude": false,
          "id": "123e4567-e89b-12d3-a456-426614174000",
          "mountPath": "/mnt/data"
        }
      ],
      "emptyDirVolume": [
        {
          "exclude": false,
          "medium": "string",
          "name": "storage-instance-a",
          "path": "/mnt/emptydir",
          "sizeLimit": "1G"
        }
      ],
      "git": [
        {
          "branch": "main",
          "exclude": false,
          "name": "storage-instance-a",
          "passwordSecret": "my-password-secret",
          "path": "/container/my-repository",
          "repository": "https://github.com/my-git/my-repo",
          "revision": "string",
          "secretKeyOfPassword": "Password",
          "secretKeyOfUser": "User",
          "secretRef": {
            "name": "my-password-secret",
            "authenticationMethod": "password",
            "secretKeyOfUser": "User",
            "secretKeyOfPassword": "Password"
          }
        }
      ],
      "hostPath": [
        {
          "exclude": false,
          "mountPath": "/local/directory",
          "mountPropagation": "None",
          "name": "storage-instance-a",
          "path": "/container/directory",
          "readOnly": true
        }
      ],
      "nfs": [
        {
          "exclude": false,
          "mountPath": "/local/nfs",
          "name": "storage-instance-a",
          "path": "/container/nfs",
          "readOnly": true,
          "server": "my.nfs.com"
        }
      ],
      "pvc": [
        {
          "claimInfo": {
            "size": "1G",
            "storageClass": "my-storage-class",
            "accessModes": {
              "readWriteOnce": true,
              "readOnlyMany": false,
              "readWriteMany": false
            },
            "volumeMode": "Filesystem",
            "addedAttrValues": [
              {
                "key": "dnsname",
                "value": "my.dns.com"
              }
            ]
          },
          "claimName": "my-claim",
          "dataSharing": false,
          "ephemeral": false,
          "exclude": false,
          "existingPvc": false,
          "name": "storage-instance-a",
          "path": "/container/my-claim",
          "readOnly": false
        }
      ],
      "s3": [
        {
          "accessKeySecret": "my-access-key-secret",
          "bucket": "my-bucket",
          "exclude": false,
          "name": "storage-instance-a",
          "path": "/container/my-bucket",
          "secretKeyOfAccessKeyId": "AccessKeyId",
          "secretKeyOfSecretKey": "SecretKey",
          "url": "https://s3.amazonaws.com"
        }
      ],
      "secretVolume": [
        {
          "defaultMode": "0644",
          "exclude": false,
          "mountPath": "string",
          "name": "storage-instance-a",
          "secret": "string"
        }
      ]
    },
    "terminateAfterPreemption": false,
    "terminationGracePeriodSeconds": 20,
    "tolerations": [
      {
        "name": "string",
        "operator": "Equal",
        "key": "string",
        "value": "string",
        "effect": "NoSchedule",
        "seconds": 1,
        "exclude": false
      }
    ],
    "tty": true,
    "workingDir": "/home/myfolder"
  },
  "masterSpecSameAsWorker": true,
  "masterSpec": {
    "annotations": [
      {
        "name": "billing",
        "value": "my-billing-unit",
        "exclude": false
      }
    ],
    "args": "-x my-script.py",
    "autoDeletionTimeAfterCompletionSeconds": 15,
    "backoffLimit": 3,
    "category": "string",
    "command": "python",
    "compute": {
      "cpuCoreLimit": 2,
      "cpuCoreRequest": 0.5,
      "cpuMemoryLimit": "30M",
      "cpuMemoryRequest": "20M",
      "extendedResources": [
        {
          "resource": "hardware-vendor.example/foo",
          "quantity": "2",
          "exclude": false
        }
      ],
      "gpuDevicesRequest": 1,
      "gpuMemoryLimit": "10M",
      "gpuMemoryRequest": "10M",
      "gpuPortionLimit": 0.5,
      "gpuPortionRequest": 0.5,
      "gpuRequestType": "portion",
      "largeShmRequest": false
    },
    "createHomeDir": true,
    "environmentVariables": [
      {
        "name": "HOME",
        "value": "/home/my-folder",
        "secret": {
          "name": "postgress_secret",
          "key": "POSTGRES_PASSWORD"
        },
        "configMap": {
          "name": "my-config-map",
          "key": "MY_POSTGRES_SCHEMA"
        },
        "podFieldRef": {
          "path": "metadata.name"
        },
        "userCredential": {
          "name": "my_postgres_user_and_password",
          "key": "POSTGRES_PASSWORD"
        },
        "exclude": false,
        "description": "Home directory of the user."
      }
    ],
    "exposedUrls": [
      {
        "container": 8080,
        "url": "https://my-url.com",
        "authorizationType": "authenticatedUsers",
        "authorizedUsers": [
          "user-a",
          "user-b"
        ],
        "authorizedGroups": [
          "group-a",
          "group-b"
        ],
        "toolType": "jupyter",
        "toolName": "my-pytorch",
        "name": "url-instance-a",
        "exclude": false
      }
    ],
    "image": "python:3.8",
    "imagePullPolicy": "Always",
    "imagePullSecrets": [
      {
        "name": "string",
        "userCredential": true,
        "exclude": false
      }
    ],
    "labels": [
      {
        "name": "stage",
        "value": "initial-research",
        "exclude": false
      }
    ],
    "nodeAffinityRequired": {
      "nodeSelectorTerms": [
        {
          "matchExpressions": [
            {
              "key": "string",
              "operator": "In",
              "values": [
                "string"
              ]
            }
          ]
        }
      ]
    },
    "nodePools": [
      "my-node-pool-a",
      "my-node-pool-b"
    ],
    "nodeType": "my-node-type",
    "podAffinity": {
      "type": "Required",
      "key": "string"
    },
    "ports": [
      {
        "container": 8080,
        "serviceType": "LoadBalancer",
        "external": 30080,
        "toolType": "pytorch",
        "toolName": "my-pytorch",
        "name": "port-instance-a",
        "exclude": false
      }
    ],
    "preemptibility": "preemptible",
    "priorityClass": "string",
    "probes": {
      "readiness": {
        "initialDelaySeconds": 1,
        "periodSeconds": 1,
        "timeoutSeconds": 1,
        "successThreshold": 1,
        "failureThreshold": 1,
        "handler": {
          "httpGet": {
            "path": "/",
            "port": 1,
            "host": "example.com",
            "scheme": "HTTP"
          }
        }
      }
    },
    "relatedUrls": [
      {
        "url": "https://my-url.com",
        "type": "wandb",
        "name": "url-instance-a",
        "exclude": false
      }
    ],
    "restartPolicy": "Always",
    "security": {
      "allowPrivilegeEscalation": false,
      "capabilities": [
        "CHOWN",
        "KILL"
      ],
      "hostIpc": false,
      "hostNetwork": false,
      "readOnlyRootFilesystem": false,
      "runAsGid": 30,
      "runAsNonRoot": true,
      "runAsUid": 500,
      "seccompProfileType": "RuntimeDefault",
      "supplementalGroups": "2,3,5,8",
      "uidGidSource": "fromTheImage"
    },
    "stdin": true,
    "storage": {
      "configMapVolume": [
        {
          "configMap": "string",
          "defaultMode": "0644",
          "exclude": false,
          "mountPath": "string",
          "name": "storage-instance-a",
          "subPath": "string"
        }
      ],
      "dataVolume": [
        {
          "exclude": false,
          "id": "123e4567-e89b-12d3-a456-426614174000",
          "mountPath": "/mnt/data"
        }
      ],
      "emptyDirVolume": [
        {
          "exclude": false,
          "medium": "string",
          "name": "storage-instance-a",
          "path": "/mnt/emptydir",
          "sizeLimit": "1G"
        }
      ],
      "git": [
        {
          "branch": "main",
          "exclude": false,
          "name": "storage-instance-a",
          "passwordSecret": "my-password-secret",
          "path": "/container/my-repository",
          "repository": "https://github.com/my-git/my-repo",
          "revision": "string",
          "secretKeyOfPassword": "Password",
          "secretKeyOfUser": "User",
          "secretRef": {
            "name": "my-password-secret",
            "authenticationMethod": "password",
            "secretKeyOfUser": "User",
            "secretKeyOfPassword": "Password"
          }
        }
      ],
      "hostPath": [
        {
          "exclude": false,
          "mountPath": "/local/directory",
          "mountPropagation": "None",
          "name": "storage-instance-a",
          "path": "/container/directory",
          "readOnly": true
        }
      ],
      "nfs": [
        {
          "exclude": false,
          "mountPath": "/local/nfs",
          "name": "storage-instance-a",
          "path": "/container/nfs",
          "readOnly": true,
          "server": "my.nfs.com"
        }
      ],
      "pvc": [
        {
          "claimInfo": {
            "size": "1G",
            "storageClass": "my-storage-class",
            "accessModes": {
              "readWriteOnce": true,
              "readOnlyMany": false,
              "readWriteMany": false
            },
            "volumeMode": "Filesystem",
            "addedAttrValues": [
              {
                "key": "dnsname",
                "value": "my.dns.com"
              }
            ]
          },
          "claimName": "my-claim",
          "dataSharing": false,
          "ephemeral": false,
          "exclude": false,
          "existingPvc": false,
          "name": "storage-instance-a",
          "path": "/container/my-claim",
          "readOnly": false
        }
      ],
      "s3": [
        {
          "accessKeySecret": "my-access-key-secret",
          "bucket": "my-bucket",
          "exclude": false,
          "name": "storage-instance-a",
          "path": "/container/my-bucket",
          "secretKeyOfAccessKeyId": "AccessKeyId",
          "secretKeyOfSecretKey": "SecretKey",
          "url": "https://s3.amazonaws.com"
        }
      ],
      "secretVolume": [
        {
          "defaultMode": "0644",
          "exclude": false,
          "mountPath": "string",
          "name": "storage-instance-a",
          "secret": "string"
        }
      ]
    },
    "terminateAfterPreemption": false,
    "terminationGracePeriodSeconds": 20,
    "tolerations": [
      {
        "name": "string",
        "operator": "Equal",
        "key": "string",
        "value": "string",
        "effect": "NoSchedule",
        "seconds": 1,
        "exclude": false
      }
    ],
    "tty": true,
    "workingDir": "/home/myfolder"
  }
}

Use to create a distributed training.

Authentication

AuthorizationBearer

Bearer authentication

Request

This endpoint expects an object.

namestringRequiredformat: ".*">=1 character

The name of the workload.

projectIdstringRequiredformat: ".*"

The id of the project.

clusterIdstringRequiredformat: "uuid"

The id of the cluster.

useGivenNameAsPrefixbooleanOptionalDefaults to false

When true, the requested name will be treated as a prefix. The final name of the workload will be composed of the name followed by a random set of characters.

templateIdstring or nullOptionalformat: "uuid"

The unique identifier of the template to use for submitting this workload. The combined values provided in the spec, template and assets will be used to create the workload.

specobjectOptional

masterSpecSameAsWorkerbooleanOptional

used for distributed workloads to indicate that the master spec should be the same as the worker spec. in this case, masterSpec should not be specified.

masterSpecobjectOptional

Response

Request completed successfully.

namestringformat: ".*">=1 character

The name of the workload.

requestedNamestring

The name as was requested for the workload. If useGivenNameAsPrefix, in the creation request, is false, name and requestedName should be identical. Otherwise, name should be composed of requestedName followed by a suffix of random characters.

workloadIdstringformat: "uuid"

A unique ID of the workload.

projectIdstringformat: ".*"

The id of the project.

clusterIdstringformat: "uuid"

The id of the cluster.

createdBystring

The user who created the workload

createdAtdatetime

The creation time of the workload.

desiredPhaseenum

The desired phase of the workload.

Allowed values:

departmentIdstringOptionalformat: ".*">=1 character

The id of the department.

deletedAtdatetime or nullOptional

The deletion time of the workload.

actualPhaseenumOptional

specobjectOptional

masterSpecSameAsWorkerbooleanOptional

used for distributed workloads to indicate that the master spec should be the same as the worker spec. in this case, masterSpec should not be specified.

masterSpecobjectOptional

Errors

400

Bad Request Error

401

Unauthorized Error

403

Forbidden Error

503

Service Unavailable Error