Create and manage custom storage tables

Create a custom storage table

1. Log on to the IoT Platform console.

2. On the Instance Overview page, find the target Enterprise instance and click the instance to open the Instance Details page.

3. In the navigation pane on the left, choose DataService Studio > Data Storage.

4. On the Data Storage page, click the Offline Storage tab.

5. On the Offline Storage tab, click Custom Storage Tables.

   

6. On the Custom Storage Tables tab, click Create Custom Storage Table.

7. In the Create Custom Storage Table dialog box, you can configure the following parameters.

   Parameter	Example	Required	Description
   Table Display Name	Daily Average Temperature Storage Table	Yes	The name of the storage table. The name cannot start with a digit or an underscore (_). It can contain Chinese characters, letters, digits, and underscores (_). The name can be up to 30 characters in length.
   Table Identifier	daily_average_temperature	Yes	The identifier of the storage table. The identifier must start with a letter. It can contain letters, digits, and underscores (_). The identifier can be up to 30 characters in length.
   Table Type	Partitioned table	Yes	Optional: Partitioned table: Suitable for storing data from timed scheduling. You can create a partition every hour or every day. When the corresponding task is rerun, the data in the entire partition is overwritten. Important If your custom storage table generates more than 30 million rows of data for each scheduling, select Partitioned table to ensure query performance. Transactional table: Similar to a table in a relational database such as MySQL. This table type is suitable for custom primary keys, where data is updated and overwritten at the row level based on the primary key granularity. Important This table can store a maximum of 30 million rows of data. Data that exceeds this limit cannot be written. Time series table: Suitable for storing data that is continuously appended over time, such as data reported by devices or events triggered by devices.
   Partition Granularity	Day	No	If you set Table Type to Partitioned table and enable automatic partitioning, you must configure the partition granularity. You can set it to Hour or Day. Important This setting must be consistent with the scheduling period in the SQL analysis task. For more information, see Set a task scheduling policy.
   Description	Average temperature of the thermometer	No	The purpose of the storage table.
   Data Storage Duration	More	Yes	The period for which data is stored in the table. The system deletes data that is older than the specified period based on the time field in the table. You can select 1 Month, 2 Months, 3 Months, 6 Months, 1 Year, Never Expire, or More. If you select More, you must specify Time Unit and Time Value.
   Time Unit	Year	No	The time unit for data storage. This parameter is required if you set Data Storage Duration to More. Valid values: Year, Month, and Day.
   Time Value	10	No	The time value for data storage. This parameter is required if you set Data Storage Duration to More. The value must be a positive integer. The maximum values are: 102 (Year). 2142 (Month). 64260 (Day).
   Data backflow to IoT Digital Twin		No	Select the Data backflow to IoT Digital Twin check box and then select an associated digital twin space. You can select only Enterprise Edition or Ultimate Edition IoT Digital Twin spaces under the current instance. Important If you only have the Trial Edition of the IoT Digital Twin engine, this option is unavailable (grayed out), and you must purchase the Enterprise or Ultimate Edition of the IoT Digital Twin engine. For more information, see Purchase an Enterprise or Ultimate Edition. Once selected, the data in this table can be used as a data source for digital twin entities in the IoT digital twin space. For more information, see Configure a data source. The new table must contain the following columns: value: The value is of the double type. iot_id: The iotId of a twin node in a twin entity. property_id: The identifier of a property in a twin node. After the data table is created: You cannot clear the Data backflow to IoT Digital Twin check box or change the associated digital twin space. You cannot add new fields to the table.

8. Click Confirm to create the custom storage table.

   After the table is created, you can use it to store parsed data.

   For more information, see Data parsing overview.

Manage custom storage tables

View table information

1. On the Custom Storage Tables tab, the storage list displays basic information for each storage table, including Table Display Name, Table Type, Partition Granularity, Table Identifier, Storage Usage, and Description.

2. Find the storage table that you created and click View in the Actions column to view its details.

Item	Example	Description
Table Identifier	${system.device_group_relation}	The identifier of the storage table.
Table Display Name	Device Group Relation Table	The Chinese name of the storage table.
Number of Fields	8	The number of items in the table header.
Data Source	System data source	The data sources are as follows: System data source: The data comes from a platform system table. Product property data source, Product event data source: The data comes from a time series table or a snapshot table.
Storage Duration	Never Expire	The data in the storage table never expires.
Backflow to IoT Digital Twin	Yes	The data in the storage table flows back to the specified digital twin space in IoT Digital Twin.
Reflowed Twin Space	Campus Digital Twin
Partitioned	Yes	Automatic partitioning is enabled for the storage table data. The partition granularity is Day.
Partition Granularity	Day
Description	This table describes the relationship between devices and groups	The purpose of the storage table.
Flink Task		This configuration item is displayed for time series tables and is disabled by default. You can enable a Flink task to integrate the time series data into Alibaba Cloud Realtime Compute for Flink for computation and analysis. For more information, see the Manage time series table data integration (Flink) section of this topic.

Define table schema

1. On the Custom Storage Tables tab, find the storage table that you created and click View in the Actions column.

2. On the Storage Table Details page, click the Table Schema tab.

   The Table Schema tab displays partition details for the storage table (such as whether it is partitioned, the partition key name, partition key type, and partition key granularity) and details for each field (such as the field name, field type, and whether it is a primary key).

   • Partitioned table: By default, it includes the partition_key field (a system-assigned partition key that cannot be changed) and the ts field (a system-assigned timestamp column that cannot be changed).

   • Transactional table: By default, it includes the ts field (a system-assigned timestamp column that cannot be changed).

   • Time series table: The table schema is empty by default.

3. Click Add Field, configure the table header fields, and then click Apply.

   The following table describes the field parameters.

Parameter	Description
Field Name	It can contain lowercase letters, digits, and underscores (_). It cannot start with a digit or an underscore (_). It can be up to 32 characters in length.
Field Display Name	It can contain Chinese characters, letters, digits, and underscores (_). It cannot start with a digit or an underscore (_). It can be up to 64 characters in length.
Field Type	Valid values: VARCHAR, BIGINT, BOOLEAN, TIMESTAMP, and DOUBLE.
Is Primary Key	Specifies whether the field is a primary key.
Remarks	The purpose of the field.

4. Repeat the previous step to add multiple fields to the storage table.

5. Click Save and then click Confirm in the dialog box that appears.

   

Optional operations

On the Custom Storage Tables tab, find your storage table in the list and click an action in the Actions column.

Manage time series table data integration (Flink)

Enable a Flink task

You can enable a Flink task for a time series table to integrate its data into Realtime Compute for Flink. This lets you perform real-time computation and analysis to diagnose device status, detect faults, and predict product yield.

1. In the storage list on the Custom Storage Tables tab, find the time series table that you created and click View in the Actions column.

2. On the storage table details page, click the shutdown icon on the Flink Job switch.

   

3. In the dialog box that appears, click Confirm to Enable.

   

Disable a Flink task

If you no longer need to use Realtime Compute for Flink for data stream computation and analysis, you can disable the Flink task.

1. On the storage table details page, turn on the Flink Job switch.

2. Click Confirm to Disable.