Cross-tenant publishing migrates data between tenants. This feature supports various data application scenarios and helps businesses maximize the value of their data.
Prerequisites
Purchase the cross-tenant publishing value-added service and enable the cross-tenant publishing module for the current tenant. If you plan to export or import label architecture, labels, or quality objects, you must enable these features first.
Before you publish across tenants, ensure that the source and target tenants use the same Dataphin version and compute engine type. Otherwise, publishing compute source objects will fail. For more information about compute engine settings, see Compute settings.
Confirm that the source and target tenants use the same cluster as the compute engine. If the compute engines are different, you must first migrate the compute engine's metadata and business data.
Ensure that the target tenant has network connectivity to the compute engine and the business database.
Ensure that both the source and target tenants have cross-tenant publishing maintenance mode enabled.
The source and target tenants must use the same cross-tenant publishing credentials.
Limits
For custom data sources, you must manually create the target environment and ensure that the data source type encoding is consistent between the source and target environments. Inconsistencies can cause unknown errors during publishing.
NoteCross-tenant publishing is not supported for custom data sources created in Dataphin versions earlier than v3.10.
Configuration files for data sources and compute sources cannot be exported. You must upload them again to the target environment.
You cannot export Fusioninsight Flink real-time compute sources.
You cannot export real-time computing templates or real-time nodes created from templates.
DataService Studio APIs cannot be exported. You must migrate them manually.
Ad hoc query files cannot be exported.
During cross-tenant publishing, users who are not system administrators or cross-tenant publishing users can only access the following menus:
Data Development & Construction > Development (Data Integration, Data Studio, Task Publishing, Task O&M)
Asset Governance & Management > Administration (Asset Checklist)
Asset Analysis & Application > Analysis (SQL Query, Notebook)
More > Management Center (Permission Management)
More > My (Personal Center, Personal Data Center, Alert Center, Message Center, Task Hub)
Custom menus
Background information
Dataphin completely isolates data between tenants to ensure data security. To build a data ecosystem or enforce strict security controls, such as creating common development, test, or production environments across different tenants, you would typically have to redevelop all data in the new tenant. This process reduces the efficiency of your data applications.
To address this issue, Dataphin has enhanced its data publishing capabilities to provide a simple, efficient, and secure platform for cross-tenant publishing.
Cross-tenant publishing is suitable for the following situations:
Environment migration (instance migration), such as migrating from an on-premises deployment to a public cloud Software as a Service (SaaS) environment.
Isolating development and production environments (instances), such as separating production and development in the finance industry.
Function Overview
Cross-tenant publishing supports batch publishing of objects from a file. The system automatically determines the publishing order based on object dependencies to ensure that they are published to the target environment correctly. Dataphin supports cross-tenant publishing for objects related to planning, data architecture, development, and assets.
Global objects for tenant publishing: Includes statistical periods, global variables, public calendars, offline scheduling templates, data segments, compute sources, projects, data sources, object attributes, and detection features.
Data architecture objects for cross-tenant publishing: Includes subject areas, business objects, and business activities.
Development objects for cross-tenant publishing: Includes integration nodes, modeling development objects, metric development objects, resource functions, and real-time and offline computing nodes.
Label architecture objects for cross-tenant publishing: Includes label entities and label entity IDs.
Label objects for cross-tenant publishing: Includes offline labels (metric mapping, rule combinations) and offline views.
Data Standard objects for cross-tenant publishing: Includes standard objects (data standard set folders, data standard sets, standard templates, data standards, public standard properties), standard mapping rules, mapping relationships, lookup tables, and root words.
Data Quality objects for cross-tenant publishing: Includes rule templates and quality rules.
Data security objects for cross-tenant publishing: Includes security rules (data classification, data categorization, detection features), detection results, and keys.
Glossary
Tenant: In a Dataphin deployment instance, the space can be isolated into tenants, which can be further divided into business segments and projects. Cross-tenant publishing is the process of exporting development objects from a tenant in the source environment into a deployment package, and then importing the exported deployment package into a tenant in the target environment.
Source environment: The environment from which data is migrated to another environment.
Target environment: The environment to which objects are published. The source environment and target environment are relative concepts. For example, if data from environment A needs to be published to environment B, then environment A is the source environment and environment B is the target environment.
Cross-tenant publishing user: A user role that has permissions for the cross-tenant publishing feature.
Procedure
Main flow | Description |
Step 1: Prepare for cross-tenant publishing |
|
Step 2: Configure cross-tenant publishing | The cross-tenant publishing user configures settings in the source and target environments. These settings include the retention period for deployment package files, obtaining or setting credentials, permission verification for features and data, and flow settings. For more information, see Cross-tenant publishing settings. |
Step 3: Export and download the deployment package from the source environment | In the source environment, the cross-tenant publishing user exports and downloads the deployment package file. You can export a deployment package by performing a full export, an incremental export, or a specified object export.
Note The exported deployment package does not include instance data generated by Operations and Maintenance (O&M). |
Step 4: Import the deployment package into the target environment | In the target environment, the cross-tenant publishing user imports the deployment package file. For more information, see Import a deployment package. |
Step 5: Publish objects in the target environment | In the target environment, select the imported deployment package and publish the objects as needed. For more information, see Objects to be published. |
Notes on cross-tenant publishing
During cross-tenant publishing, you can create new nodes in both the source and target tenants, but you cannot publish them to the production environment. Nodes cannot be submitted in the Basic environment. If you create new nodes in a draft or development state during this process, you must manually migrate them to sync them to the target tenant.
Dataphin generates a global dependency graph based on data lineage. It uses this graph to determine the correct publishing order and ensure the accuracy of dependencies. When you use the one-click publishing feature, you do not need to manage the publishing order.
Before you use one-click publishing, you must manually publish the compute sources and data sources.
During cross-tenant publishing, the system automatically generates a global ID for each object. This ensures that each object can be uniquely identified when published across different environments.