Common job identifier conventions - MaxCompute - Alibaba Cloud Documentation Center

This topic describes the common identifier conventions for MaxCompute jobs.

Introduction to parameter identifiers

MaxCompute job identifier conventions ensure that information about jobs submitted from upper-layer platforms, such as DataWorks, is recorded in the job metadata. This information includes the node ID, name, and owner ID of the submitting platform. MaxCompute parses and records these parameter identifiers from each job's settings.

By default, jobs submitted from DataWorks and DataPhin include the following parameter identifiers.

Field Name	Value	Description
EXT_PLATFORM_ID	Job source platform	Identifies the upper-layer platform that submitted the MaxCompute job. The same platform must use a consistent identifier. Length limit: 32
EXT_NODE_ID	Upper-layer scheduling node ID	Identifies the node ID of the upper-layer application that submitted the MaxCompute job. Length limit: 64
EXT_DAGTYPE	Instance running mode	0: Normal scheduling instance 1: One-time task 2: Smoke testing 3: Data backfill
EXT_TASK_ID	Upper-layer scheduling node instance ID	Identifies the specific running instance of EXT_NODE_ID.
EXT_NODE_NAME	Name of the upper-layer scheduling node	Identifies the name of the upper-layer application node that submitted the MaxCompute job. Length limit: 128
EXT_NODE_ONDUTY	ID of the upper-layer scheduling node owner	Identifies the owner ID of the upper-layer application. Length limit: 64
EXT_NODE_PRIORITY	Upper-layer scheduling node priority	Identifies the priority of the upper-layer node. Numeric format. Note that settings is a string-to-string map.

Submit parameter identifiers

The following example shows how to identify an SQL job upon submission by adding a flag that defines its source SDK, name (for example, example_task), and owner (for example, test_user):

SET EXT_PLATFORM_ID='SDK';
SET EXT_NODE_NAME='example_task';
SET EXT_NODE_ONDUTY ='1234567';
sqlquery;

Use parameter identifiers

This information is recorded in the tasks_history view of the tenant-level Information Schema. This makes it easier to associate jobs with an upper-layer platform for analysis and statistics.
Search in the console
1. Log on to the MaxCompute console and select a region in the top-left corner.
2. In the navigation pane on the left, choose Observation O&M > Jobs.
3. The job list displays the ExtPlatformId, ExtNodeId, ExtNodeName, and ExtNodeOnDuty information for each job. You can search by ExtNodeId or ExtNodeName to quickly find a job and identify its source node ID, name, and owner.