The scheduling period defines how often a node runs automatically. In a production environment, the scheduling system uses this period to run the node. Recurring tasks generate recurring instances based on the schedule type and period. The system then runs these instances to execute the tasks automatically.
Usage notes
A node's scheduling frequency is independent of the periods of its upstream nodes.
A node runs based on its own scheduling period, not the period of its upstream nodes.
DataWorks supports dependencies between nodes with different scheduling periods.
In DataWorks, recurring tasks generate recurring instances based on the schedule type and period. For example, a task with an hourly period generates a corresponding number of hourly instances each day. Nodes run as instances. The dependencies that you set for recurring tasks are essentially dependencies between the instances that are generated by the nodes. If upstream and downstream nodes have different scheduling periods, the number of recurring instances and their dependencies will differ. For more information about upstream and downstream dependencies for different scheduling periods, see Principles and examples for configuring complex dependency scheduling.
Nodes dry-run outside their scheduled time.
In DataWorks, nodes that are not scheduled to run every day, such as weekly or monthly nodes, will dry-run on their non-scheduled days. When the scheduled time for a dry-run instance is reached, the node immediately returns a 'Success' status. This allows a downstream daily scheduled node to be triggered and run normally.
Node execution time.
This setting only specifies the expected schedule time for the node. The actual execution time is affected by multiple factors, such as the scheduled time of upstream nodes, the availability of execution resources, and the runtime conditions of the node.
When you update a node's schedule time, if you set Instance Generation Method to Generate instance on the next day (T+1) and republish the node to the Operation Center, the scheduled times of instances that have run in the last two days will change based on the new schedule time. Instances that have not yet run will use the new schedule time.
Instance Generation Method to Generate instance now, new instances are generated based on the new schedule time. The scheduled times of historical instances do not change.
Use cases for timed scheduling
If all nodes in a business flow must start at a specific time, you can set the start time only for the first node in the flow. The downstream nodes will run in sequence according to their dependencies after the first node runs. For more information about how dependencies affect node execution, see How dependencies affect node execution.
If nodes in a business flow have different start time requirements, you can set the schedule time for each node individually.
Examples
In a business flow, Node C depends on Node B, and Node B depends on Node A. Node A is the first node in the business flow. The following examples provide more details.
Diagram | Description |
 | The entire business flow needs to start at 3:00. You only need to set the schedule time for Node A to 3:00. Although the default schedule time for the downstream nodes is 0:00, they will wait for Node A to finish before they start running. |
 | Node A is scheduled to run at 3:00. However, Node B must run at 5:00, and Node C must run at 6:00. In this case, you must set separate schedule times for Node B and Node C. |
 | Node A is scheduled to run at 3:00, but Node B must run at 5:00. Node C uses the default configuration and waits for Node B to finish. Therefore, Node C runs after 5:00. |
Scheduling methods
DataWorks supports minute, hourly, daily, weekly, monthly, and yearly scheduling. The configuration and runtime details for each method are described below.
Minute scheduling
The minimum interval for minute scheduling is 1 minute.
Configuration example
Configuration path
Go to the node editing page in Data Development. In the right-side navigation pane, click Scheduling. In the Scheduling Time section, configure the node's scheduling period.
Example
The target node is scheduled to run every 30 minutes from
00:00to23:59every day. The configuration is shown in the following figure.NoteThe cron expression is automatically generated based on your selections and cannot be manually edited.
Scheduling details
The following figure shows the scheduled times and parameter substitution for instances of a node that is scheduled to run every 30 minutes.
For more information about dependency use cases for minute scheduling, see Minute task dependencies.
Hourly scheduling
Notes
The time period is calculated based on a closed interval [start, end]. For example, if you configure a task to run every hour from 00:00 to 03:00, the time range is [00:00, 03:00] with a 1-hour interval. The scheduling system generates four instances each day, with scheduled runtimes at 00:00, 01:00, 02:00, and 03:00.
You can set a node to run at a specified interval within a specific time range each day. You can also select multiple points in time for the node to run.
The time points configured for a recurring schedule are the scheduled times. The actual runtime may differ from the scheduled time due to factors such as resource availability.
Configuration example
Configuration path
Go to the node editing page in Data Development. In the right-side navigation pane, click Scheduling. In the Scheduling Time section, configure the node's scheduling period.
Configuration details.
The target node is scheduled to run automatically every 6 hours from
00:00to23:59every day. The configuration is shown in the following figure.NoteThe cron expression is automatically generated based on your selections and cannot be manually edited.
Scheduling details
The scheduling system generates four instances each day. The instances run at their scheduled times of 00:00, 06:00, 12:00, and 18:00, as shown in the following figure.
For more information about dependency scenarios for hourly scheduling, see Hourly task dependencies.
Daily scheduling
Daily scheduling means that a scheduled node runs once a day at a specified time. When you create a recurring task, the default schedule time is a random time between 00:00 and 00:30. You can specify a different runtime as needed. For example, you can specify the node to run at 13:00 every day.
Configuration example
Configuration path
Go to the node editing page in Data Development. In the right-side navigation pane, click Scheduling. In the Scheduling Time section, configure the node's scheduling period.
Configuration details
Assume that the import, statistical processing, and export nodes are all daily scheduled nodes.
The runtime for these nodes is
13:00every day.The statistical processing node depends on the import node, and the export node depends on the statistical processing node.
Based on this scenario, the configuration for the daily scheduled nodes is shown in the following figure.
NoteThe cron expression is automatically generated based on your selections and cannot be manually edited.
Scheduling details
The scheduling system automatically generates and runs instances for the nodes. The data processing times are shown in the following figure.
A scheduled node runs only if the following conditions are met:
The upstream nodes have run successfully.
The scheduled runtime of the node has been reached.
The node cannot run if either condition is not met. These conditions can be met in any order.
The default schedule time is a random time between
00:00and00:30.For more information about dependency scenarios for daily scheduling, see Daily task dependencies.
Weekly scheduling
Notes
On non-scheduled days, a weekly scheduled node generates dry-run recurring instances to ensure that downstream instances run as expected.
NoteDry-run behavior: The system directly returns a success status without actually running the task.
The instance status is 'Success', the execution duration is 0 seconds, and no execution logs are generated.
Scheduling resources are not consumed.
Downstream nodes that depend on the dry-run node are not blocked. This means that if a node dry-runs, its downstream daily, hourly, or minute-level tasks still run as scheduled.
The scheduling period is configured for each individual node. Whether a node runs on a given day depends on its own scheduling period, not the period of its upstream nodes. However, the scheduled time of upstream nodes affects the actual start time of downstream nodes.
Configuration example
Configuration path
Go to the node editing page in Data Development. In the right-side navigation pane, click Scheduling. In the Scheduling Time section, configure the node's scheduling period.
Configuration details.
The target node is configured to run on Mondays and Fridays. Instances generated on these days are scheduled and run normally. Instances generated on Tuesdays, Wednesdays, Thursdays, Saturdays, and Sundays will dry-run. This means that when the scheduled runtime is reached on these days, the instance status is immediately set to 'Success' without executing the code. The configuration is shown in the following figure.
NoteThe cron expression is automatically generated based on your selections and cannot be manually edited.
Scheduling details
The scheduling system automatically generates and runs instances for the nodes.
When you use the data backfill feature for a weekly scheduled node, note that the date you select for backfilling is the business date. Business date = Scheduled date - 1 day.
For example:
For a node that runs every Monday, you must select the previous Sunday as the business date for data backfill.
If you select a business date other than Sunday, the data backfill instance will dry-run.
For more information about complex dependency scenarios, see Principles and examples for configuring complex dependency scheduling.
Monthly scheduling
Notes
On non-scheduled days, a monthly scheduled node generates dry-run recurring instances to ensure that downstream instances run as expected.
NoteDry-run behavior: The system directly returns a success status without actually running the task.
The instance status is 'Success', the execution duration is 0 seconds, and no execution logs are generated.
Scheduling resources are not consumed.
Downstream nodes that depend on the dry-run node are not blocked. This means that if a node dry-runs, its downstream daily, hourly, or minute-level tasks still run as scheduled.
The scheduling period is configured for each individual node. Whether a node runs on a given day depends on its own scheduling period, not the period of its upstream nodes. However, the scheduled time of upstream nodes affects the actual start time of downstream nodes.
For monthly scheduling, you can set the Specified Time to the Last Day Of Each Month. The node will then run on the last day of every month.
Configuration example
Configuration path
Go to the node editing page in Data Development. In the right-side navigation pane, click Scheduling. In the Scheduling Time section, configure the node's scheduling period.
Configuration details
The target node is configured to run on the last day of each month. The instance generated on the last day of the month is scheduled and run normally. Instances generated on other days will dry-run. This means that when the scheduled runtime is reached on these days, the instance status is immediately set to 'Success' without executing the code. The configuration is shown in the following figure.
NoteThe cron expression is automatically generated based on your selections and cannot be manually edited.
Scheduling details
The scheduling system automatically generates and runs instances for the nodes, as shown in the following figure.
When you use the data backfill feature for a monthly scheduled task, the date you select for backfilling is treated as the business date. Business date = Scheduled date - 1 day.
For example:
For a node that runs on the first day of each month, you must select the last day of the previous month as the business date for data backfill.
For a node that runs on the last day of each month, you must select the day before the last day of the month as the business date.
If you select any other business date, the data backfill instance will dry-run.
For more information about complex dependency scenarios, see Principles and examples for configuring complex dependency scheduling.
Yearly scheduling
Notes
On non-scheduled days, a yearly scheduled node generates dry-run recurring instances to ensure that downstream instances run as expected.
Dry-run behavior: The system directly returns a success status without actually running the task.
The instance status is 'Success', the execution duration is 0 seconds, and no execution logs are generated.
Scheduling resources are not consumed.
Downstream nodes that depend on the dry-run node are not blocked. This means that if a node dry-runs, its downstream daily, hourly, or minute-level tasks still run as scheduled.
Configuration example
Configuration path
Go to the node editing page in Data Development. In the right-side navigation pane, click Scheduling. In the Scheduling Time section, configure the node's scheduling period.
Configuration details
The target node is configured to run on the first and the last day of January, April, July, and October each year. The instances generated on these specified dates are scheduled and run normally. Instances generated on other dates will dry-run. This means that when the scheduled runtime is reached on these days, the instance status is immediately set to 'Success' without executing the code. The configuration is shown in the following figure.
Scheduling details
The scheduling system automatically generates and runs instances for the nodes, as shown in the following figure.
For more information about complex dependency scenarios, see Principles and examples for configuring complex dependency scheduling.