DataWorks:Scenario: Change the time zone for scheduling

Regions that support changing of the time zone for scheduling

DataWorks allows you to change the time zone for scheduling in workspaces that reside in specific regions. The following table describes the regions that support changing of the time zone for scheduling and the supported time zones for each region.

(Required) Precautions for changing the time zone for scheduling

To prevent errors from occurring on nodes in the production environment and unexpected running results of nodes due to the change in the time zone for scheduling, make sure that you are familiar with the precautions before you change the time zone for scheduling. The following table describes the precautions.

Impacts that are exerted on specific items in DataWorks after you change the time zone for scheduling

1. Scheduling time of a node

2. Settings of time-related parameters of a node

The scheduling parameters of a node use the related time information, such as the scheduling time and data timestamp of the node, as base values and replace the variables in the node code by using the base values. The base values vary based on the change in the time zone for scheduling.

YYYYMMDD=${yyyymmdd} LAST_2D=${yyyymmdd-2}

3. Processing logic for time information in the code of a node

Some variables in the code of a node are parsed by gateway nodes, and the rest of the variables in the node code are parsed at the related compute engine side. Variable parsing logic:

• For variables in the code of a node that are parsed by gateway nodes, the values of the variables vary based on the time zone to which you change the current time zone. In a scheduling scenario, if you directly run a command, such as the date command for a Shell node, to query a point in time in a DataWorks node, such as a Shell node or a PyODPS node, the system returns a point in time based on the time zone for scheduling. Temporarily running a node in DataStudio is not a scheduling scenario.

• For variables in the code of a node that are parsed at the related underlying compute engine side, the values of the variables are not affected by the change in the time zone for scheduling. We recommend that you refer to the time zone conversion rules of the underlying compute engine for details. For more information, see the documentation about the time zone of the underlying compute engine.

The following figure shows an example on how a variable in the code of a Hive node is replaced with a time string based on the scheduling time of the node and how the time string is issued to the Hive compute engine in a scheduling scenario. The specific time that is expressed is limited by the parsing logic at the Hive server side.

4. Time information related to baselines or alerts

The time information related to baselines or alerts varies based on the time zone to which you change the current time zone.

5. Time information related to APIs

The time information related to APIs varies based on the time zone to which you change the current time zone.

6. Time zone of an underlying engine

The time zone of an underlying engine, such as Data Integration or a MaxCompute compute engine, is determined by the underlying engine and is irrelevant to the change in the time zone for scheduling. The replacement values of scheduling parameters of a node are issued to the related underlying engine as strings for processing based on the processing rules at the engine side. For information about the processing rules at the engine side, see the documentation about the time zone of the engine.

How to change the time zone for scheduling

DataWorks allows you to change the current time zone to another time zone for scheduling of nodes in DataWorks.

Scenario 1: The current tenant creates a workspace in a region for the first time.

The first time the current tenant creates a workspace in a region that supports changing of the time zone for scheduling, a message prompts the tenant to specify whether to change the time zone for scheduling to a time zone other than the local time zone. The tenant can determine whether to change the time zone for scheduling based on business requirements and the descriptions in (Required) Precautions for changing the time zone for scheduling. If the tenant wants to change the time zone for scheduling, the tenant must submit a ticket for technical support.

Scenario 2: The current tenant creates a workspace in a region where existing workspaces of the tenant reside.

When the current tenant creates a workspace in a region where existing workspaces of the tenant reside and the region supports changing of the time zone for scheduling, the tenant can also submit a ticket to change the time zone for scheduling for all workspaces in the region. However, migration of existing nodes and instances may be involved and the business may be affected when the tenant performs the change. Therefore, the tenant must fully understand the consequences of the change. The change roughly involves the following aspects:

1. User side: Evaluate the scope of impacts of the change on existing nodes and instances.

2. DataWorks side: Migrate existing nodes and instances.

3. DataWorks side: Specify a new time zone.

4. User side: Restore existing nodes and instances and verify the new time zone.

   We recommend that you check each type of node multiple times to confirm that different types of nodes can run as expected.

Check whether the time zone for scheduling is changed

You can go to the Properties tab of a desired node in DataStudio to check whether the time zone for scheduling is changed and view the current time zone for scheduling.

1. Go to the DataStudio page.

   Log on to the DataWorks console. In the left-side navigation pane, choose Data Modeling and Development > DataStudio. On the page that appears, select the desired workspace from the drop-down list and click Go to DataStudio.

2. Go to the Properties tab.

   1. On the DataStudio page, find the desired node and go to the configuration tab of the node.

   2. On the configuration tab of the node, click Properties in the right-side navigation pane.

3. On the Properties tab, view the current time zone for scheduling in the Schedule section.

Appendix: Processing of time information related to Data Integration

The time zone of Data Integration is irrelevant to the time zone for scheduling in DataWorks. The replacement values of scheduling parameters of a node are issued to Data Integration as strings. For example, gmt_modify >= ${yyyymmdd} in the code of a node indicates that the SQL statement with the WHERE clause used for data filtering is executed to issue string literals to the related data source for processing. The data filtering effect is limited by the time zone processing mechanism of the data source.