Offline mode scheduling configuration - Dataphin - Alibaba Cloud Documentation Center

This topic describes how to set up scheduling configurations for stream-batch integrated tasks, which dictate the recurring schedule of nodes within the production environment. You can establish the task's scheduling period and effective date by adjusting the configuration properties.

Procedure

Navigate to the Dataphin home page, and from the top menu bar, select Development > Data Development.
At the top menu bar of the Development page, choose Project.
In the left-side navigation pane, click on Data Processing > Script Task. Then, in the Script Task list, select the desired task name.
On the task tab, access the Configuration panel by clicking Configuration on the right.

Within the Configuration panel, select the Offline Mode tab. Proceed to the Scan Configuration section to input the necessary parameter information.

Parameter	Description
Time Attribute	Select Time Attribute. The Time Attribute options include the following: Normal: Configure scheduling based on the scheduling period and execute normally. This option is usually selected by default for tasks. Dry-Run: Configure scheduling based on the scheduling period, but all executions are dry-run. That is, the task is marked as successfully completed as soon as it is scheduled without actually being executed.
Priority	The priority of the current real-time task in offline mode can be set to Medium, Low, or Lowest. When creating a new real-time task and configuring offline mode, the default value of this priority is derived from the default priority in Management Center > Development Platform Settings > Node Task Settings. Note When the task is published to the production environment or submitted in the Basic environment, the priority cannot be modified during task editing. It must be modified in the production environment's operations. At this time, the priority value is the latest value in the production environment.
Paused	If Paused is set to Yes, the scheduling of this task will be paused. The task will be scheduled according to the configured scheduling period, but once it is scheduled, it will directly return as failed without execution. This is typically used in scenarios where a task is temporarily not needed but will be used again later.
Recurrence	The scheduling period defines how often the task's code will be executed in the production environment scheduling system. After a task is successfully submitted, the scheduling system generates recurring instances starting from the next day based on the task's time attribute configuration. These instances are executed according to the results of upstream dependencies and the specified time points. The Scheduling Period options include Daily, Weekly, Monthly, Hourly, and Minute: Daily Scheduling: The task is automatically executed once a day. When creating a new recurring task, the system's default period is set to execute at 00:00 every day. You can specify the execution time by clicking the icon. Weekly Scheduling: The task is automatically executed once on specific days of the week at specific times. You can specify the execution time by clicking the icon. Note To ensure the normal operation of downstream instances outside the specified scheduling time, the system generates instances and directly marks them as successfully executed without performing any logic or consuming resources. Monthly Scheduling: The task is automatically executed once on specific days of the month at specific times. You can specify the execution time by clicking the icon. Note To ensure the normal operation of downstream instances outside the specified scheduling time, the system generates instances daily and directly marks them as successfully executed without performing any logic or consuming resources. Hourly Scheduling: Supports Time Range, On the Hour, Custom Time Range, and Custom Time Points scheduling. Time Range Scheduling: The task is automatically scheduled at specified intervals within a specified time range each day. Click the icon to set the Start Time and End Time of the time range. Then, click the icon to select the Interval (n hours). For example, if the time range is set to 00:00–23:00 and the interval is set to 1 hour, the task will be automatically scheduled every hour within the time range of 00:00–23:00 each day. On the Hour Scheduling: Select the On The Hour time from the drop-down list. The scheduling system will automatically generate and execute instances for the task at the selected hourly time. Custom Time Points Scheduling: The scheduling system automatically generates and executes instances for the task at the selected custom time points. Select custom time points from the drop-down list. You can click + Add Custom Time Point to add multiple time points. Custom Time Range Scheduling: The task is automatically scheduled at specified intervals within certain time ranges each day. Click the icon to set the Start Time and End Time of the time range. Then, enter the Interval (n minutes) (n is an integer between 5 and 360). You can click + Add Custom Time Range to add up to 10 time ranges. The time ranges must not overlap. For example, if Time Range 1 is set to 00:00–03:00 with an interval of 30 minutes, the task will be automatically scheduled every 30 minutes within the time range of 00:00–03:00 each day. Minute Scheduling: Supports Daily or Hourly. Daily: The scheduling task runs once at the specified time interval within the daily time period. Hourly: From the specified start time each hour to the end of that hour, the scheduled task runs once at the interval specified by the interval time number.
Scheduling Execution Plan	Click Preview. The scheduling execution plan displays all scheduling instances and their types for each day of a selected month based on the configured scheduling period and conditions. The preview date type can be selected as Data Timestamp or Execution Date (Scheduling Date). If multiple scheduling types exist for all instances on a single day, the system displays all scheduling types by color and lists the name and count of each scheduling type. For example, the following figure shows that on the 4th day of a certain month, the current scheduling task has 44 normal scheduling instances, 2 paused instances, and 12 dry-run instances. Hover the mouse over the scheduling type module for a specific day to view the detailed scheduling instance list for that day. The list includes scheduling types, scheduling conditions, and condition names.
Depend On Previous Period	Based on the business scenario, determine whether the execution of the current period's node depends on the execution results of the previous period's node or other nodes. Select the node type. The system supports selecting Custom or Current. The applicable scenarios are as follows: If the execution of the current period's node depends on whether the previous period's node successfully produces data, select Current. The current node will only start execution if the previous period's node is successfully executed. If a code task does not use the output table of a certain node but requires dependency on whether the node successfully produces data in the previous period, select Custom node dependency.

To finalize the scheduling configuration for the real-time task in offline mode, click OK.