Dataphin:Create a data plate

A data plate is a key component of a logical space and serves as a namespace organized based on business characteristics. Before you use standard modeling, you must create a data plate.

Prerequisites

• This feature is available only in the Intelligent R&D Edition. To use it, you must also enable the Intelligent R&D module for the current tenant.

• Before you begin, make sure you have added the data plate architect, business owner, and data owner to the Dataphin member list. For more information, see Add a Dataphin member.

Permissions

Only a super administrator or system administrator can create a data plate.

Overview

Dataphin supports two data plate modes: Dev-Prod mode and Basic mode.

• Dev-Prod mode: Isolates Dev and Prod environments to enhance data security in the Prod environment. Recommended if you have strong governance controls, a large data development team with well-defined roles, and a higher budget for compute and storage.

• Basic mode: Provides a single, integrated environment for development and operations, ensuring a stable and streamlined data production process. Recommended if you prioritize development efficiency, have developers with flexible roles, and have a limited budget for compute and storage.

Procedure

1. On the Dataphin homepage, click Plan in the top navigation bar. The Data Architecture page is displayed by default.

2. On the Data Architecture page, click + Create data plate.

3. In the Create data plate wizard, configure the data plate information.

   1. Step 1: Select a deployment mode.

      1. Select a deployment mode based on your requirements.

         • Dev-Prod mode: Recommended if you require strong governance, have a large team of data developers with well-defined roles, and have a higher budget for compute and storage. In this mode, the Dev and Prod environments are logically isolated. This strengthens control over the data development process and strictly manages permissions on production data to maximize data production stability. However, this mode offers less flexibility, requires higher user proficiency with the tools, and consumes duplicate compute and storage resources.

         • Basic mode: Recommended if you prioritize development efficiency over strict controls, have developers with flexible roles, and have a limited compute and storage budget. This mode provides an integrated environment for development and operations, which enables agile management of the data development process and balances production stability with flexibility. The system strictly controls write permissions for production data. Be aware that this mode carries the risk of direct changes to the production environment. Manage project members carefully to maintain the stability of production data.

           Important

           If you select Basic Mode for the Deployment Mode, the data plate cannot be upgraded to Dev-Prod Mode after it is created.

      2. Click Next.

   2. Step 2: Configure Plate Definition.

      1. Configure the basic and business information for the data plate as described in the following table.

         Parameter	Description
         Basic Information
         Plate English Name	The English name of the data plate. The name must meet the following requirements: Can contain only letters, digits, and underscores (_). Must be no more than 64 characters in length. After the data plate is created, the English name of the data plate in the Dev environment is in the format LD_{CustomPlateEnglishName}_dev. To avoid impacting data production, you cannot edit the English name after you create logical tables in the data plate.
         Plate Name	The name of the data plate. The name must meet the following requirements: Can contain only Chinese characters, letters, digits, underscores (_), and hyphens (-). Must be no more than 64 characters in length. After the data plate is created, the name of the data plate in the Dev environment is in the format {CustomPlateName}_dev.
         Description (Optional)	Enter a brief description of the data plate. The description must not exceed 128 characters.
         Icon	The icon helps you categorize the data plate by business area and does not affect data computation. You can choose from icons including E-commerce ( icon), Finance ( icon), Cloud Computing ( icon), Ad & Marketing ( icon), Logistics ( icon), Entertainment ( icon), Travel ( icon), Health ( icon), Social & Communication ( icon), Catering ( icon), Education ( icon), and Environment ( icon).
         Business Information
         Data Plate Architect	Manages data plate settings, including basic and business information. Select any active member from the Dataphin member list.
         Business Owner (Optional)	Responsible for data stability in the data plate. Select up to three members from the list.
         Data Owner (Optional)	Responsible for data quality in the data plate. Select up to three members from the list.

      2. Click Next.

   3. Step 3: Configure Logical Table Naming Convention.

      When you create a logical table, the system automatically suggests a name based on the naming convention. You can edit the suggested name.

      Note

      After you create the data plate, you can edit the naming convention by navigating to Development Specification > Table Specification > Logical Table Naming Convention for the data plate. Modifying a table prefix affects all subtypes within the same logical table category. For example, modifying the dimension logical table category affects subtypes such as regular dimension logical tables and hierarchical dimension logical tables.

      The default naming conventions and suffixes are as follows:

      Logical Table Type	Default
      Dimension Logical Table	Table prefix: dim_. Data timeliness (suffix): Regular dimension logical table: Day (T+1): None. Hour (T+h): _hf. Minute (T+m): _mf. Hierarchical dimension logical table: Day (T+1): _lvl[n], where [n] represents the number of levels. Enumerated dimension logical table and Virtual dimension logical table: Day (T+1): None.
      Fact Logical Table	Table prefix: fct_. Data timeliness (suffix): Process fact logical table: Day (T+1): None. Event fact logical table: Day (T+1): _di. Hour (T+h): _hi. Minute (T+m): _mi. Snapshot fact logical table: Day (T+1): _df. Hour (T+h): _hf. Minute (T+m): _mf. Subject Domain Abbreviation: Disabled by default. If enabled, the subject domain abbreviation is automatically added to the table name.
      Summary Logical Table	Table prefix: dws_. Data timeliness (suffix): Day (T+1): None. Hour (T+h): _hh. Minute (T+m): _mm.

4. Click OK to create the data plate.

Multi-engine environments

In a multi-engine environment, you must also select an engine type during the plate definition step. Available engine types correspond to clusters created in cluster management. Other configurations are the same as in a single-engine environment. You cannot change the engine type after the data plate is created.

Next steps

After you create the data plate, create a subject domain for it. For more information, see Create a subject domain.