Cross-tenant publishing facilitates data migration between tenants, supporting various data application scenarios and enhancing the value of enterprise data applications.
Prerequisites
You must purchase the cross-tenant publishing value-added service and activate the module for the current tenant. To export or import tag architecture, tags, or quality objects during cross-tenant publishing, enable the relevant features first.
Before you publish data across tenants, ensure that the source and target tenants have the same Dataphin version and compute engine type. This prevents errors when you publish compute source objects. For more information about compute engine settings, see Compute settings.
Verify whether the target and source tenants use the same cluster as the compute engine. If the compute engines differ, you must migrate metadata and business data first.
Ensure network connectivity between the target tenant, compute engine, and business databases.
Confirm that both the target and source tenants have enabled cross-tenant publishing maintenance mode.
The target and source tenants must use identical cross-tenant publishing credentials.
Limits
You must manually create custom data sources in the target environment and ensure data source type encoding consistency. Inconsistencies may cause unknown exceptions during publishing.
NoteCustom data sources from Dataphin versions earlier than v3.10 are not supported for cross-tenant publishing.
Configuration files associated with data sources and compute sources cannot be exported and must be re-uploaded when publishing.
For real-time compute sources, only Flink VVP compute sources support export. Other real-time compute sources do not support export.
Data service APIs cannot be exported and require manual migration.
Ad hoc query files cannot be exported.
During cross-tenant publishing, users other than system administrators and cross-tenant publishing users can only access the following menus:
Data Development > Development (Data Integration, Data Development, Task Publishing, Task O&M)
Asset Governance > Administration (Asset Checklist)
Asset Analytics > Analytics (SQL Query, Notebook)
More > Management Hub (Permission Management)
More > My (Personal Center, Personal Data Center, Alerting Center, Message Hub, Task Hub)
Custom menus
Background information
Dataphin ensures data security by isolating data between tenants. When building a data ecosystem or requiring stringent security controls, such as creating common development, testing, or production environments across different tenants, redeveloping data in a new tenant is usually necessary. This can reduce the efficiency of data applications.
To address this, Dataphin has enhanced its data publishing capabilities, offering a straightforward, efficient, secure, and reliable platform for cross-tenant publishing.
Cross-tenant publishing is applicable in the following scenarios:
Environment migration (between different instances), such as transitioning from on-premises deployment to public cloud Software as a Service (SaaS).
Isolation between development and production environments (instances), such as production-development isolation in the finance industry.
Feature overview
Cross-tenant publishing allows for batch publishing of file objects, with the system automatically managing the publishing sequence based on object dependencies. Dataphin supports the cross-tenant publishing of various objects, including those related to planning, data architecture, development, and assets.
Publishing global objects across tenants: This includes statistical periods, global variables, public calendars, offline scheduling templates, data sections, compute sources, projects, data sources, object attributes, and recognition features.
Publishing data architecture-related objects across tenants: This encompasses subject areas, business objects, and business activities.
Publishing development objects across tenants: This category includes integration tasks, modeling development objects, metric development objects, resource functions, and real-time offline computing tasks.
Publishing tag architecture objects across tenants: This covers tag entities and tag entity IDs.
Publishing tag objects across tenants: This includes offline tags (metric mappings, rule combinations) and offline views.
Publishing data standard objects across tenants: This involves standard objects (standard set directories, standard sets, standard templates, data standards, public standard attributes), standard mapping rules, mapping relationships, lookup tables, and roots.
Publishing data quality objects across tenants: This includes rule templates and quality rules.
Publishing data security objects across tenants: This involves security rules (data classification, data grading, recognition features), recognition results, and keys.
Glossary
Tenant: A tenant in Dataphin's deployment instance refers to an isolated space where business sections and projects are organized. Cross-tenant publishing involves exporting development objects from a source environment tenant into a publishing file, then importing this file into a tenant in the target environment.
Source environment: The original environment from which data is migrated to other environments.
Target environment: The destination environment for published objects. Source and target environments are relative; for instance, if data from environment A is published to environment B, A is the source and B is the target.
Cross-tenant publishing user: A user role with permissions to perform cross-tenant publishing functions.
Procedure
Main procedure | Description |
Step 1: Cross-tenant publishing preparation |
|
Step 2: Cross-tenant publishing settings | Users managing cross-tenant publishing can configure the retention period for package files, manage cross-tenant publishing credentials, validate feature and data permissions, and establish procedures in both the source and target environments. For detailed instructions, see cross-tenant publishing settings. |
Step 3: Export and download publishing files in the source environment | In the source environment, users with cross-tenant publishing permissions can export and download publishing package files. This process supports full, incremental, and specified object export methods.
Note Note that the exported publishing package files do not include instance data generated by operations in the development module. |
Step 4: Import publishing files in the target environment | In the target environment, users with cross-tenant publishing permissions import the publishing package files. For detailed instructions, see import publishing package. |
Step 5: Publish objects in the target environment | In the target environment, select the imported publishing package based on your business needs and publish the required objects. For more information, see Objects to be published. |
Notes for cross-tenant publishing
During cross-tenant publishing, both source and target tenants can create new tasks but cannot publish them to the production environment (tasks cannot be submitted in the Basic environment). Any newly created draft or development tasks that need to be synchronized to the target tenant require manual migration.
Dataphin generates a global dependency graph based on data lineage relationships and uses this graph to create the correct publishing sequence, ensuring dependency accuracy. When using the one-click publishing feature, you do not need to worry about the publishing order.
Before using one-click publishing, it is recommended to manually publish compute sources and data sources first.
During cross-tenant publishing, the system automatically generates global IDs for each object to ensure unique identification across different environments.