DeadlineCloud / Client / create_fleet

create_fleet#

DeadlineCloud.Client.create_fleet(**kwargs)#

Creates a fleet. Fleets gather information relating to compute, or capacity, for renders within your farms. You can choose to manage your own capacity or opt to have fleets fully managed by Deadline Cloud.

See also: AWS API Documentation

Request Syntax

response = client.create_fleet(
    clientToken='string',
    farmId='string',
    displayName='string',
    description='string',
    roleArn='string',
    minWorkerCount=123,
    maxWorkerCount=123,
    configuration={
        'customerManaged': {
            'mode': 'NO_SCALING'|'EVENT_BASED_AUTO_SCALING',
            'workerCapabilities': {
                'vCpuCount': {
                    'min': 123,
                    'max': 123
                },
                'memoryMiB': {
                    'min': 123,
                    'max': 123
                },
                'acceleratorTypes': [
                    'gpu',
                ],
                'acceleratorCount': {
                    'min': 123,
                    'max': 123
                },
                'acceleratorTotalMemoryMiB': {
                    'min': 123,
                    'max': 123
                },
                'osFamily': 'WINDOWS'|'LINUX'|'MACOS',
                'cpuArchitectureType': 'x86_64'|'arm64',
                'customAmounts': [
                    {
                        'name': 'string',
                        'min': ...,
                        'max': ...
                    },
                ],
                'customAttributes': [
                    {
                        'name': 'string',
                        'values': [
                            'string',
                        ]
                    },
                ]
            },
            'storageProfileId': 'string'
        },
        'serviceManagedEc2': {
            'instanceCapabilities': {
                'vCpuCount': {
                    'min': 123,
                    'max': 123
                },
                'memoryMiB': {
                    'min': 123,
                    'max': 123
                },
                'osFamily': 'WINDOWS'|'LINUX',
                'cpuArchitectureType': 'x86_64'|'arm64',
                'rootEbsVolume': {
                    'sizeGiB': 123,
                    'iops': 123,
                    'throughputMiB': 123
                },
                'acceleratorCapabilities': {
                    'selections': [
                        {
                            'name': 't4'|'a10g'|'l4'|'l40s',
                            'runtime': 'string'
                        },
                    ],
                    'count': {
                        'min': 123,
                        'max': 123
                    }
                },
                'allowedInstanceTypes': [
                    'string',
                ],
                'excludedInstanceTypes': [
                    'string',
                ],
                'customAmounts': [
                    {
                        'name': 'string',
                        'min': ...,
                        'max': ...
                    },
                ],
                'customAttributes': [
                    {
                        'name': 'string',
                        'values': [
                            'string',
                        ]
                    },
                ]
            },
            'instanceMarketOptions': {
                'type': 'on-demand'|'spot'
            }
        }
    },
    tags={
        'string': 'string'
    }
)
Parameters:

  • clientToken (string) –

    The unique token which the server uses to recognize retries of the same request.

    This field is autopopulated if not provided.

  • farmId (string) –

    [REQUIRED]

    The farm ID of the farm to connect to the fleet.

  • displayName (string) –

    [REQUIRED]

    The display name of the fleet.

    Warning

    This field can store any content. Escape or encode this content before displaying it on a webpage or any other system that might interpret the content of this field.

  • description (string) –

    The description of the fleet.

    Warning

    This field can store any content. Escape or encode this content before displaying it on a webpage or any other system that might interpret the content of this field.

  • roleArn (string) –

    [REQUIRED]

    The IAM role ARN for the role that the fleet’s workers will use.

  • minWorkerCount (integer) – The minimum number of workers for the fleet.

  • maxWorkerCount (integer) –

    [REQUIRED]

    The maximum number of workers for the fleet.

  • configuration (dict) –

    [REQUIRED]

    The configuration settings for the fleet. Customer managed fleets are self-managed. Service managed Amazon EC2 fleets are managed by Deadline Cloud.

    Note

    This is a Tagged Union structure. Only one of the following top level keys can be set: customerManaged, serviceManagedEc2.

    • customerManaged (dict) –

      The customer managed fleets within a fleet configuration.

      • mode (string) – [REQUIRED]

        The Auto Scaling mode for the customer managed fleet configuration.

      • workerCapabilities (dict) – [REQUIRED]

        The worker capabilities for a customer managed fleet configuration.

        • vCpuCount (dict) – [REQUIRED]

          The vCPU count for the customer manged worker capabilities.

          • min (integer) – [REQUIRED]

            The minimum amount of vCPU.

          • max (integer) –

            The maximum amount of vCPU.

        • memoryMiB (dict) – [REQUIRED]

          The memory (MiB).

          • min (integer) – [REQUIRED]

            The minimum amount of memory (in MiB).

          • max (integer) –

            The maximum amount of memory (in MiB).

        • acceleratorTypes (list) –

          The accelerator types for the customer managed worker capabilities.

          • (string) –

        • acceleratorCount (dict) –

          The range of the accelerator.

          • min (integer) – [REQUIRED]

            The minimum number of GPUs for the accelerator. If you set the value to 0, a worker will still have 1 GPU.

          • max (integer) –

            The maximum number of GPUs for the accelerator.

        • acceleratorTotalMemoryMiB (dict) –

          The total memory (MiB) for the customer managed worker capabilities.

          • min (integer) – [REQUIRED]

            The minimum amount of memory to use for the accelerator, measured in MiB.

          • max (integer) –

            The maximum amount of memory to use for the accelerator, measured in MiB.

        • osFamily (string) – [REQUIRED]

          The operating system (OS) family.

        • cpuArchitectureType (string) – [REQUIRED]

          The CPU architecture type for the customer managed worker capabilities.

        • customAmounts (list) –

          Custom requirement ranges for customer managed worker capabilities.

          • (dict) –

            The fleet amount and attribute capabilities.

            • name (string) – [REQUIRED]

              The name of the fleet capability.

            • min (float) – [REQUIRED]

              The minimum amount of fleet worker capability.

            • max (float) –

              The maximum amount of the fleet worker capability.

        • customAttributes (list) –

          Custom attributes for the customer manged worker capabilities.

          • (dict) –

            Defines the fleet’s capability name, minimum, and maximum.

            • name (string) – [REQUIRED]

              The name of the fleet attribute capability for the worker.

            • values (list) – [REQUIRED]

              The number of fleet attribute capabilities.

              • (string) –

      • storageProfileId (string) –

        The storage profile ID.

    • serviceManagedEc2 (dict) –

      The service managed Amazon EC2 instances for a fleet configuration.

      • instanceCapabilities (dict) – [REQUIRED]

        The Amazon EC2 instance capabilities.

        • vCpuCount (dict) – [REQUIRED]

          The amount of vCPU to require for instances in this fleet.

          • min (integer) – [REQUIRED]

            The minimum amount of vCPU.

          • max (integer) –

            The maximum amount of vCPU.

        • memoryMiB (dict) – [REQUIRED]

          The memory, as MiB, for the Amazon EC2 instance type.

          • min (integer) – [REQUIRED]

            The minimum amount of memory (in MiB).

          • max (integer) –

            The maximum amount of memory (in MiB).

        • osFamily (string) – [REQUIRED]

          The operating system (OS) family.

        • cpuArchitectureType (string) – [REQUIRED]

          The CPU architecture type.

        • rootEbsVolume (dict) –

          The root EBS volume.

          • sizeGiB (integer) –

            The EBS volume size in GiB.

          • iops (integer) –

            The IOPS per volume.

          • throughputMiB (integer) –

            The throughput per volume in MiB.

        • acceleratorCapabilities (dict) –

          The GPU accelerator capabilities required for the Amazon EC2 instances. If you include the acceleratorCapabilities property in the ServiceManagedEc2InstanceCapabilities object, all of the Amazon EC2 instances will have at least one accelerator.

          • selections (list) – [REQUIRED]

            A list of objects that contain the GPU name of the accelerator and driver for the instance types that support the accelerator.

            • (dict) –

              Values that you can use to select a particular Amazon EC2 instance type.

              • name (string) – [REQUIRED]

                The name of the GPU accelerator.

              • runtime (string) –

                The driver version that the GPU accelerator uses.

          • count (dict) –

            The number of GPUs on each worker. The default is 1.

            • min (integer) – [REQUIRED]

              The minimum number of GPUs for the accelerator. If you set the value to 0, a worker will still have 1 GPU.

            • max (integer) –

              The maximum number of GPUs for the accelerator.

        • allowedInstanceTypes (list) –

          The allowable Amazon EC2 instance types.

          • (string) –

        • excludedInstanceTypes (list) –

          The instance types to exclude from the fleet.

          • (string) –

        • customAmounts (list) –

          The custom capability amounts to require for instances in this fleet.

          • (dict) –

            The fleet amount and attribute capabilities.

            • name (string) – [REQUIRED]

              The name of the fleet capability.

            • min (float) – [REQUIRED]

              The minimum amount of fleet worker capability.

            • max (float) –

              The maximum amount of the fleet worker capability.

        • customAttributes (list) –

          The custom capability attributes to require for instances in this fleet.

          • (dict) –

            Defines the fleet’s capability name, minimum, and maximum.

            • name (string) – [REQUIRED]

              The name of the fleet attribute capability for the worker.

            • values (list) – [REQUIRED]

              The number of fleet attribute capabilities.

              • (string) –

      • instanceMarketOptions (dict) – [REQUIRED]

        The Amazon EC2 market type.

        • type (string) – [REQUIRED]

          The Amazon EC2 instance type.

  • tags (dict) –

    Each tag consists of a tag key and a tag value. Tag keys and values are both required, but tag values can be empty strings.

    • (string) –

      • (string) –

Return type:

dict

Returns:

Response Syntax

{
    'fleetId': 'string'
}

Response Structure

  • (dict) –

    • fleetId (string) –

      The fleet ID.

Exceptions

  • DeadlineCloud.Client.exceptions.AccessDeniedException

  • DeadlineCloud.Client.exceptions.InternalServerErrorException

  • DeadlineCloud.Client.exceptions.ResourceNotFoundException

  • DeadlineCloud.Client.exceptions.ThrottlingException

  • DeadlineCloud.Client.exceptions.ValidationException

  • DeadlineCloud.Client.exceptions.ServiceQuotaExceededException