Automatically generated based on the configured period and execution time.
Note：
A one-off task cannot be used to generate new instances via resubmission after completion. You need to create a new task to repeat execution.
Minute Scheduling
Parameter description:
Parameter
Description
Configuration Method
A single task configuration is selected by default as normal and cannot be modified.
Scheduling Cycle
The execution cycle unit of task scheduling supports minute, hour, day, week, month, year, and one-time scheduling.
Effective Time
The effective time period of the scheduling time configuration. The system will perform auto-scheduling within this time range according to the time configuration. It will not auto-schedule after exceeding the validity period. Only select a start date after the current date. By default, it displays the current date to 2099-12-31.
Execution Time
The execution time for minute-based tasks can only be selected on the hour, with interval ranges: 5, 10, 15, 20, 30 minutes. Minute-based tasks generate instances for scheduling based on the execution time.
Calendar Scheduling
Use the configuration in the scheduling calendar to arrange task running or pause running on the specified date.
Scheduling Plan
Automatically generated based on the configured period and execution time.
Hourly Scheduling
Parameter description:
Parameter
Description
Configuration Method
A single task configuration is selected by default as normal and cannot be modified.
Scheduling Cycle
The execution cycle unit of task scheduling supports minute, hour, day, week, month, year, and one-time scheduling.
Effective Time
The effective time period of the scheduling time configuration. The system will perform auto-scheduling within this time range as per the time configuration. It will not auto-schedule after exceeding the validity period. Only select a start date.
Execution Time (choose one with the specified time)
Execution Time
Hourly task execution time is selectable in Hour:Minute format, with an interval range of 1-23 hours. Tasks generate instances for scheduling based on the execution time.
Interval Time
Set the interval time as the measurement unit to control the generation frequency of instance of tasks within the designated execution time interval.
Specified Time (choose one with the execution time)
Specified Time
Specify a time range from 0 to 23, support multiple selections. Hourly tasks generate instances and execute based on the specified time.
Specified Minute
Specify the exact minutes to generate an instance after the specified time.
Calendar Scheduling
Use the configuration in the scheduling calendar to arrange task running or pause running on the specified date.
Scheduling Plan
Automatically generated based on the configured period and execution time.
Daily Scheduling
Parameter description:
Parameter
Description
Configuration Method
Supports general and crontab configuration modes. A single task configuration is selected by default as normal and cannot be modified.
Scheduling Cycle
The execution cycle unit of task scheduling supports minute, hour, day, week, month, year, and one-time scheduling.
Effective Time
The effective time period of the scheduling time configuration. The system will perform auto-scheduling within this time range according to the time configuration. It will not auto-schedule after exceeding the validity period. Only select a start date after the current date. By default, it displays the current date to 2099-12-31.
Execution Time
Daily task execution time is selectable in Hour:Minute format. Instances start execution when reaching the execution time.
Calendar Scheduling
Use the configuration in the scheduling calendar to arrange task running or pause running on the specified date.
Scheduling Plan
Automatically generated based on the configured period and execution time.
Weekly Scheduling
Parameter description:
Parameter
Description
Configuration Method
Supports general and crontab configuration modes. A single task configuration is selected by default as normal and cannot be modified.
Scheduling Cycle
The execution cycle unit of task scheduling supports minute, hour, day, week, month, year, and one-time scheduling.
Effective Time
The effective time period of the scheduling time configuration. The system will perform auto-scheduling within this time range according to the time configuration. It will not auto-schedule after exceeding the validity period. Only select a start date after the current date. By default, it displays the current date to 2099-12-31.
Specify Date
The specified date range for weekly tasks is Monday to Sunday, supporting multiple selections.
Execution Time
Weekly task execution time is selectable in Hour:Minute format. Instances start execution when reaching the execution time on the specified date.
Calendar Scheduling
Use the configuration in the scheduling calendar to arrange task running or pause running on the specified date.
Scheduling Plan
Automatically generated based on the configured period and execution time.
Monthly Scheduling
Parameter description:
Parameter
Description
Configuration Method
Supports general and crontab configuration modes. A single task configuration is selected by default as normal and cannot be modified.
Scheduling Cycle
The execution cycle unit of task scheduling supports minute, hour, day, week, month, year, and one-time scheduling.
Effective Time
The effective time period of the scheduling time configuration. The system will perform auto-scheduling within this time range according to the time configuration. It will not auto-schedule after exceeding the validity period. Only select a start date after the current date. By default, it displays the current date to 2099-12-31.
Specify Date
The specified date range for monthly tasks is from the 1st to the 31st of each month, supporting multiple selections. Note: the end of month option is mutually exclusive with other dates.
Execution Time
Monthly task execution time is selectable in Hour:Minute format. Instances start execution when reaching the execution time on the specified date.
Scheduling Plan
Automatically generated based on the configured period and execution time.
Annual Scheduling
Parameter description:
Parameter
Description
Configuration Method
Supports general and crontab configuration modes. A single task configuration is selected by default as normal and cannot be modified.
Scheduling Cycle
The execution cycle unit of task scheduling supports minute, hour, day, week, month, year, and one-time scheduling.
Effective Time
The effective time period of the scheduling time configuration. The system will perform auto-scheduling within this time range according to the time configuration. It will not auto-schedule after exceeding the validity period. Only select a start date after the current date. By default, it displays the current date to 2099-12-31.
Specify Month
The specified month range for annual tasks is from January to December, supporting multiple selections.
Specify Date
The specified date range for annual tasks is from the 1st to the 31st, supporting multiple selections. Note: the end of month option is mutually exclusive with other dates.
Execution Time
Annual task execution time is selectable in Hour:Minute format. Instances start execution when reaching the execution time on the specified date in the specified month.
Scheduling Plan
Automatically generated based on the configured period and execution time.
Scheduling Type
Supports selecting normal scheduling and dry-run scheduling. Dry-run scheduling marks the instance as successful without actually executing the task when scheduling to the task. Normal scheduling means executing the task code normally.
Scheduling Dependency Configuration
Dependency configuration
Dependency configuration supports setting upstream and downstream dependencies and task instance execution order for task nodes.
Parameter description:
Information
Description
Self-dependent
Self-dependent (dependency between instances in the same task)
Parallel processing
No dependency between the previous and next cycle instances. If a task has multiple instances, they run simultaneously.
Unordered serial
No dependency between the current instance and the previous cycle instance. If a task has multiple instances, randomly select an instance to run. Only one instance is in the running state at a time.
Ordered serial
The current instance depends on the status of the previous cycle instance.
Workflow self-dependency
Task scheduling without workflow self-dependency configuration.
upstream dependency task
search upstream dependency task
Support through searching task name, task ID, or project name to find upstream dependency tasks that need to be configured.
upstream dependency task list
When adding searched tasks as dependencies, the dependency task list displays the corresponding task information, including task name, status, scheduling cycle, owner, associated workflow, associated project, time dimension, instance scope, execution policy, and delete.
The detailed parameter descriptions are as follows:
Task name: task name, task id, task type of the dependency task.
Task status: scheduling running status of the dependency task.
Scheduling cycle: scheduling cycle of the dependent task.
Owner: owner of the dependency task.
Associated workflow: data workflow the dependent task belongs to.
Project of the task: project the dependent task belongs to.
Configuration mode: Recommendation strategy: includes commonly used dependency configuration methods; Custom: user custom dependency configuration as needed.
Time dimension Instance range
If you select the recommendation strategy in the configuration mode: select the dependency method and scope, with different values for tasks in different cycles. For detailed configuration items, see the scheduling dependency note in the table below.
If you select custom in the configuration mode:
Interval mode: The input format is x,y, which means the range of time offsets for dependency on upstream task instance data. For example, in the case of a day interval, inputting -10,-1 means the closed interval of instances from 10 days to 1 day before the upstream task. x<=y, and the maximum difference is 1000.
List mode: Input format: x,y,z. It means the specific offset value for dependency on upstream task instance data. For example, in the case of a list (day), if you input -3,-2,-1, it means the instances from 3, 2, and 1 day before the upstream task. x<y<z, and a maximum of 50 specific time offset values can be entered.
Execution policy: Processing policy when dependent upstream task instance does not exist, defaults to wait for upstream. Select wait for upstream means persistently wait when the upstream task instance does not exist. Select execute normally means the task can operate normally in three corresponding statuses: upstream task instance does not exist, upstream task failed, or upstream timeout.
Operation
Dependency preview: Supports previewing dependency results based on configured dependency relationships to assist users in judging whether the configured dependencies meet expectations. Available start scheduling time can be selected to check the corresponding instance data time and dependent instance list (instance data time).
Delete: Changes take effect in production status only after submission upon successful deletion.
Dependency auto-parsing
Provides dependency auto-parsing to help users quickly configure dependency relationships. The specific method is as follows.
Step one: Underwrite the output registration of the upstream task is configured, see below for the method.
Step 2: Click Auto-parse in the current task.
Step 3: Select the task to depend on.
Dependency on Downstream Task
The downstream task dependency means the current task can select tasks from all downstream (including indirect downstream) to depend on, meeting users' special dependency scenarios.
For example: Task A is the upstream task of Task B. The periodic instance B1 depends on the periodic instance A1, and A1 wants to depend on B2. This means the configuration shows A depends on B, and B depends on A. You can configure Task B in the task scheduling - upstream task dependency of Task A.
Circular dependency can exist between tasks, but not between instances. Therefore, the system will periodically detect the dependency relationship of the instance. If a circular dependency on the instance is found after the task is launched, an alarm will be sent to you via email and SMS.
Scheduling dependency note
Period
Upstream
Minute
Hour
Day
Week
Month
Year
one-time
Downstream
Minute
(1) By minute/current minute: instances dependent on the same data time.
(2) By minute/previous period: instances dependent on the previous data time.
(1) By hour/current hour: ALL minute-based tasks within an hour have to wait until the hourly task operation is completed before they can run.
(1) By day/current: all minutely tasks within a day can run only after the daily task is completed.
(1) By week/this week: all minutely tasks within a week can run only after the weekly task is completed.
(1) By month/this month: all minutely tasks within a month can run only after the monthly task is completed.
(1) By year/this year: all minutely tasks within the year can run only after the annual task is completed.
All minutely instances depend on one-time task instances, irrespective of data time.
Hour
(1) By hour/previous hour (-60,0]: depends on all upstream instances in the previous hour (-60,0].
(2) By hour/previous hour: depends on all upstream instances in the previous hour [-60,0).
(3) By hour/current hour: depends on ALL upstream minute instances in the current hour.
(1) Hourly/most recent instance: dependent on the instance of the last data time. First check [0:00 on the day, current system time], if not found then check [current system time, 0:00 the next day).
(1) By day/current day: all hourly tasks within a day can run only after the daily task is completed.
(1) By week/this week: all hourly tasks within a week can run only after the weekly task is completed.
(1) By month/this month: all hourly tasks within a month can run only after the monthly task is completed.
(1) By year/this year: all hourly tasks within the year can run only after the annual task is completed.
All hourly instances depend on one-time task instances, irrespective of data time.
Day
(1) By day/previous day (-24*60,0]: depends on all upstream instances in the previous day (-24*60,0].
(2) By day/previous day: depends on all upstream instances in the previous day [-24*60,0).
(3) By day/on the day: depends on all upstream minute instances on the day.
(1) By day/previous day (-24*60,0]:
(2) By day/previous day:
(3) By day/on the day: depends on all upstream hourly instances on the day.
(1) By day/on the day: depends on the upstream daily instance on the day.
(1) By week/this week: all day tasks within a week can run only after the weekly task is completed.
(2) By day/instance of the last data time: depends on the current time and the instance of the last data time before.
(1) By month/this month: all daily tasks within a month can run only after the monthly task is completed.
(2) By day/instance of the last data time: depends on the current time and the instance of the last data time before.
(1) By year/this year: all daily tasks within the year can run only after the annual task is completed.
(2) By day/instance of the last data time: depends on the current time and the instance of the last data time before.
All daily instances depend on one-time task instances, irrespective of data time.
Week
(1) By week/last week: dependent on all minutely task instances within last week.
(2) By week/this week: dependent on all minutely tasks within this week.
(1) By week/last week: dependent on all hourly task instances within last week.
(2) By week/this week: dependent on all hourly task instances within this week.
(1) By week/last week: dependent on all daily task instances within last week.
(2) By week/last Friday: dependent on daily task instances of last Friday.
(3) By week/last Sunday: dependent on daily task instances of last Sunday.
(4) By week/this week: dependent on all daily task instances within this week.
(5) By day/on the day: dependent on daily task instances on the day.
(6) By day/previous day: dependent on daily task instances of the previous day.
(1) By week/this week: dependent on all daily task instances within this week.
(2) By day/instance of the last data time: depends on the current time and the instance of the last data time before.
(1) By month/this month: all weekly tasks within a month can run only after the monthly task is completed.
(2) By day/instance of the last data time: depends on the current time and the instance of the last data time before.
(1) By year/this year: all weekly tasks within the year can run only after the annual task is completed.
All weekly instances depend on one-time task instances, irrespective of data time.
Month
(1) By month/last month: dependent on all minute-level task instances within last month.
(2) By month/this month: dependent on all minute-level task instances within this month.
(1) By month/last month: dependent on all hourly task instances within last month.
(2) By month/this month: dependent on all hourly task instances within this month.
(1) By month/last month: dependent on all daily task instances within last month.
(2) By month/end of last month: dependent on the instance of the last day within last month.
(3) By month/this month: dependent on all daily task instances within this month.
(4) By day/on the day: dependent on daily task instances on the day.
(5) By day/previous day: dependent on daily task instances of the previous day.
(1) By month/last month: dependent on all weekly task instances within last month.
(2) By month/this month: dependent on all weekly task instances within this month.
(3) By day/instance of the last data time: depends on the current time and the instance of the last data time before.
(1) By month/this month: dependent on all monthly task instances within this month.
(2) By day/instance of the last data time: depends on the current time and the instance of the last data time before.
(1) By year/this year: all tasks within the year can run only after the annual task is completed.
All monthly instances depend on one-time task instances, irrespective of data time.
Year
(1) By year/this year: dependent on all minute-level task instances within the year.
(1) By year/this year: dependent on all hourly task instances within the year.
(1) By day/all days this year: dependent on all daily task instances within the year.
(2) By day/on the day: depends on the daily instance with the same data time.
(3) By day/previous day: depends on the instance of the previous day's data time.
(1) By year/this year: dependent on all weekly task instances within the year.
(1) By month/all months this year: dependent on all monthly task instances within the year.
(2) By month/this month: dependent on all monthly task instances within this month.
(3) By month/last month: dependent on all monthly task instances within last month.
(4) By month/end of last month: dependent on the monthly task instance of the last day within last month.
(5) By month/beginning of last month: dependent on the monthly task instance of the first day within last month.
(1) By year/this year: dependent on all upstream annual task instances within this year.
All annual instances depend on one-time task instances, unrelated to data time.
one-time
(1) By day/on the day:
All minutely task instances within the specified date.
(1) By day/on the day: Default cross-cycle: depend on all hourly task instances on the day of the specified time.
(1) By day/on the day: depend on daily task instances within the specified date.
(1) By week/this week: one-time tasks within a week can only run after the weekly task is completed.
(1) By month/this month: one-time tasks within a month can only run after the monthly task is completed.
(1) By year/this year: one-time tasks within a year can only run after the annual task is completed.
Depend on one-time task instances, irrespective of data time.
Output Registration
The current task supports configuration of generated data tables and provides two setting methods:
Manual addition: Click Add, fill in the output data table database, output data table, and output table partition (partition is optional).
Auto-parsing: Click Auto-parsing, the system will parse the output table information from the current code and fill it.
