CreateJob - Platform For AI - Alibaba Cloud Documentation Center

Creates a job that runs in a cluster. You can configure the data source, code source, startup command, and computing resources of each node on which a job runs.

Operation description

Before you call this operation, make sure that you understand the billing methods and pricing of Deep Learning Containers (DLC) of Platform for AI (PAI).

Debugging

You can run this interface directly in OpenAPI Explorer, saving you the trouble of calculating signatures. After running successfully, OpenAPI Explorer can automatically generate SDK code samples.

Debug

Authorization information

The following table shows the authorization information corresponding to the API. The authorization information can be used in the Action policy element to grant a RAM user or RAM role the permissions to call this API operation. Description:

Operation: the value that you can use in the Action element to specify the operation on a resource.
Access level: the access level of each operation. The levels are read, write, and list.
Resource type: the type of the resource on which you can authorize the RAM user or the RAM role to perform the operation. Take note of the following items:
- For mandatory resource types, indicate with a prefix of * .
- If the permissions cannot be granted at the resource level, All Resources is used in the Resource type column of the operation.
Condition Key: the condition key that is defined by the cloud service.
Associated operation: other operations that the RAM user or the RAM role must have permissions to perform to complete the operation. To complete the operation, the RAM user or the RAM role must have the permissions to perform the associated operations.

Operation	Access level	Resource type	Condition key	Associated operation
paidlc:CreateJob	create	All Resources ``	none	none

Request syntax

POST /api/v1/jobs HTTP/1.1

Request parameters

Parameter	Type	Required	Description	Example
body	object	No	The request parameters.
DisplayName	string	Yes	The job name. The name must be in the following format: The name must be 1 to 256 characters in length. The name can contain digits, letters, underscores (_), periods (.), and hyphens (-).	tf-mnist-test
JobType	string	Yes	The job type. The value is case-sensitive. The following job types are supported: TFJob PyTorchJob MPIJob XGBoostJob OneFlowJob ElasticBatchJob SlurmJob RayJob Valid values and corresponding frameworks: OneFlowJob: OneFlow. PyTorchJob: PyTorch. SlurmJob: Slurm. XGBoostJob: XGBoost. ElasticBatchJob: ElasticBatch. MPIJob: MPIJob. TFJob: Tensorflow. RayJob: Ray.	TFJob
JobSpecs	array	Yes	JobSpecs describes the configurations for job running, such as the image address, startup command, node resource declaration, and number of replicas. A DLC job consists of different types of nodes. If nodes of the same type have exactly the same configuration, the configuration is called JobSpec. JobSpecs specifies the configurations of all types of nodes. The value is of the array type.
	JobSpec	No	The runtime configurations of the job.
UserCommand	string	Yes	The startup command for all nodes of the job.	python /root/code/mnist.py
DataSources	array<object>	No	The data sources for job running.
	object	No	All data sources of the job. The data source is mounted to the local path of the container in which each node resides based on the data source configuration. The local path is specified by the MountPath parameter in the data source configuration. The process in the startup command of the job directly accesses the distributed file system that resides in the path specified by the MountPath parameter. Each data source is a distributed file system.
DataSourceId	string	No	The data source ID.	d-cn9dl*******
MountPath	string	No	The path to which the job is mounted. By default, the mount path in the data source configuration is used. This parameter is optional.	/root/data
Uri	string	No	The data source path.	oss://bucket.oss-cn-hangzhou-internal.aliyuncs.com/path/
Options	string	No	The mount attribute of the custom dataset. Set the value to OSS.	{ "fs.oss.download.thread.concurrency": "10", "fs.oss.upload.thread.concurrency": "10", "fs.jindo.args": "-oattr_timeout=3 -oentry_timeout=0 -onegative_timeout=0 -oauto_cache -ono_symlink" }
CodeSource	object	No	The code source of the job. Before the node of the job runs, DLC automatically downloads the configured code from the code source and mounts the code to the local path of the container.
CodeSourceId	string	No	The ID of the code source.	code-20210111103721-xxxxxxx
Branch	string	No	The branch of the referenced code repository. By default, the branch configured in the code source is used. This parameter is optional.	master
Commit	string	No	The commit ID of the code to be downloaded. By default, the commit ID configured in the code source is used. This parameter is optional.	44da109b5******
MountPath	string	No	The path to which the job is mounted. By default, the mount path configured in the data source is used. This parameter is optional.	/root/data
UserVpc	object	No	The VPC settings.
VpcId	string	No	The VPC ID.	vpc-abcdef****
SwitchId	string	No	The vSwitch ID. This parameter is optional. If you leave this parameter empty, the system automatically selects a vSwitch based on the inventory status. You can also specify a vSwitch ID.	vs-abcdef****
SecurityGroupId	string	No	The ID of the security group.	sg-abcdef****
ExtendedCIDRs	array	No	The extended CIDR block. If you leave the SwitchId and ExtendedCIDRs parameters empty, the system automatically obtains all CIDR blocks in a VPC. If you configure the SwitchId and ExtendedCIDRs parameters, we recommend that you specify all CIDR blocks in a VPC.
	string	No	The extended CIDR block.	192.168.0.1/24
DefaultRoute	string	No	The default route. Default value: false. Valid values: eth0: The default network interface is used to access the Internet through the public gateway. eth1: The user's elastic network interface (ENI) is used to access the Internet through the private gateway. For more information about the configuration method, see Enable Internet access for a DSW instance by using a private Internet NAT gateway.	eth0
ThirdpartyLibs	array	No	The third-party Python libraries to be installed.
	string	No	The third-party Python library and its version. Example: `numpy==1.16.1`.	numpy==1.16.1
ThirdpartyLibDir	string	No	The folder in which the third-party Python library file requirements.txt is stored. Before the startup command specified by the UserCommand parameter is run on each node, DLC fetches the requirements.txt file from the folder and runs `pip install -r` to install the required package and library.	/root/code/
Envs	object	No	The environment variables.
	string	No	The environment variable in the `Key=Value` format.	ENABLE_DEBUG_MODE
JobMaxRunningTimeMinutes	long	No	The maximum running duration of the job. Unit: minutes.	1024
WorkspaceId	string	No	The workspace ID.	ws-20210126170216-xxxxxxx
ResourceId	string	No	The ID of the resource group. This parameter is optional. If you leave this parameter empty, the job is submitted to a public resource group. If a resource quota is bound to the current workspace, you can specify the resource quota ID. For more information about how to query the resource quota ID, see Manage resource quotas.	rs-xxx
Priority	integer	No	The priority of the job. Default value: 1. Valid values: 1 to 9. 1 is the lowest priority. 9: the highest priority.	8
Settings	JobSettings	No	The additional parameter configurations of the job.
ElasticSpec	JobElasticSpec	No	This parameter is not supported.
DebuggerConfigContent	string	No	This parameter is not supported.	“”
Options	string	No	The additional configuration of the job. You can use this parameter to adjust the behavior of the attached data source. For example, if the attached data source of the job is of the OSS type, you can use this parameter to add the following configurations to override the default parameters of JindoFS: `fs.oss.download.thread.concurrency=4,fs.oss.download.queue.size=16`.	key1=value1,key2=value2
SuccessPolicy	string	No	The policy that is used to check whether a distributed multi-node job is successful. Only TensorFlow distributed multi-node jobs are supported. ChiefWorker: If you use this policy, the job is considered successful when the pod on the chief node completes operations. AllWorkers (default): If you use this policy, the job is considered successful when all worker nodes complete operations.	AllWorkers
CredentialConfig	CredentialConfig	No	The access credential configuration.
Accessibility	string	No	The job visibility. Valid values: PUBLIC: The job is visible to all members in the workspace. PRIVATE: The job is visible only to you and the administrator of the workspace.	PRIVATE

Response parameters

Parameter	Type	Description	Example
	object	The response parameters.
JobId	string	The job ID.	dlc7*******
RequestId	string	The request ID used to troubleshoot issues.	473469C7-AA6F-4DC5-B3DB-xxxxxxx

Examples

Sample success responses

JSONformat

{
  "JobId": "dlc7*******",
  "RequestId": "473469C7-AA6F-4DC5-B3DB-xxxxxxx"
}

Error codes

For a list of error codes, visit the Service error codes.

Change history

Change time	Summary of changes	Operation
2024-12-18	The internal configuration of the API is changed, but the call is not affected	View Change Details
2024-08-09	The internal configuration of the API is changed, but the call is not affected	View Change Details
2024-07-05	The internal configuration of the API is changed, but the call is not affected	View Change Details
2023-12-08	The internal configuration of the API is changed, but the call is not affected	View Change Details
2023-09-11	The internal configuration of the API is changed, but the call is not affected	View Change Details