Set Up Tenants to Enforce Data Isolation - Dataphin - Alibaba Cloud - Dataphin

By creating multiple tenants, you can assign distinct Data Units and Function Modules to each. This approach ensures data isolation between tenants and helps meet their unique requirements. This topic describes how to create, edit, delete, and restore tenants.

Permissions

Only Super Administrators and System Administrators in the YuanCang Tenant can create and manage Tenants.

Engine features

Super X

Type	X-code assistant	X-O&M assistant	X-data engineering	X-analysis	X-asset Q and A	X-catalog management
MaxCompute	Supported	Supported	Supported	Supported	Supported	Supported
AnalyticDB for PostgreSQL	Supported	Supported	Supported	Supported	Supported	Supported
E-MapReduce 3.x	Supported	Supported	Supported	Supported	Supported	Supported
E-MapReduce 5.x	Supported	Supported	Supported	Supported	Supported	Supported
CDH 5.x	Supported	Supported	Supported	Supported	Supported	Supported
CDH 6.x	Supported	Supported	Supported	Supported	Supported	Supported
Cloudera Data Platform 7.x	Supported	Supported	Supported	Supported	Supported	Supported
Huawei FusionInsight 8.x	Supported	Supported	Supported	Supported	Supported	Supported
AsiaInfo DP 5.3	Supported	Supported	Supported	Supported	Supported	Supported
Transwarp Data Hub (TDH) 6.x	Supported	Supported	Supported	Supported	Supported	Supported
Transwarp Data Hub (TDH) 9.3.x	Supported	Supported	Supported	Supported	Supported	Supported
Transwarp ArgoDB	Supported	Supported	Supported	Supported	Supported	Supported
StarRocks	Supported	Supported	Supported	Supported	Supported	Supported
Lindorm (Compute Engine)	Supported	Supported	Supported	Supported	Supported	Supported
GaussDB (DWS)	Supported	Supported	Supported	Supported	Supported	Supported
Databricks	Supported	Supported	Supported	Supported	Supported	Supported
Amazon EMR	Supported	Supported	Supported	Supported	Supported	Supported
SelectDB	Supported	Supported	Supported	Supported	Supported	Supported
Doris	Supported	Supported	Supported	Supported	Supported	Supported
Aliyun EMR Serverless Spark	Supported	Supported	Supported	Supported	Supported	Supported

Development

Engine type	Intelligent development	Real-time integration	Real-time development	Intelligent O&M
MaxCompute	Supported	Supported	Supported	Supported
AnalyticDB for PostgreSQL	Supported	Supported	Supported	Supported
E-MapReduce 3.x	Supported	Supported	Supported	Supported
E-MapReduce 5.x	Supported	Supported	Supported	Supported
CDH 5.x	Supported	Supported	Supported	Supported
CDH 6.x	Supported	Supported	Supported	Supported
Cloudera Data Platform 7.x	Supported	Supported	Supported	Supported
Huawei FusionInsight 8.x	Supported	Supported	Supported	Supported
AsiaInfo DP 5.3	Supported	Supported	Supported	Supported
Transwarp TDH 6.x	Supported	Supported	Supported	Supported
Transwarp TDH 9.3.x	Supported	Supported	Supported	Supported
Transwarp ArgoDB	Supported	Supported	Not supported	Supported
StarRocks	Supported	Supported	Supported	Supported
Lindorm (Compute Engine)	Supported	Supported	Supported	Supported
GaussDB (Data Warehouse Service (DWS))	Not supported	Supported	Supported	Supported
Databricks	Supported	Supported	Supported	Supported
Amazon EMR	Supported	Supported	Supported	Supported
SelectDB	Supported	Supported	Supported	Supported
Doris	Supported	Supported	Supported	Supported
Aliyun EMR Serverless Spark	Supported	Supported	Supported	Supported

Governance

Engine type	Data standard	Asset security	Resource governance	Metadata center	Intra-domain asset quality	Global asset quality
MaxCompute	Supported	Supported	Supported	Supported	Supported	Supported
AnalyticDB for PostgreSQL	Supported	Supported	Not Supported	Supported	Supported	Supported
E-MapReduce (EMR) 3.x	Supported	Supported	Supported	Supported	Supported	Supported
E-MapReduce (EMR) 5.x	Supported	Supported	Supported	Supported	Supported	Supported
CDH 5.x	Supported	Supported	Supported	Supported	Supported	Supported
CDH 6.x	Supported	Supported	Supported	Supported	Supported	Supported
Cloudera Data Platform 7.x	Supported	Supported	Supported	Supported	Supported	Supported
Huawei FusionInsight 8.x	Supported	Supported	Supported	Supported	Supported	Supported
AsiaInfo DP 5.3	Supported	Supported	Supported	Supported	Supported	Supported
Transwarp Data Hub (TDH) 6.x	Supported	Supported	Supported	Supported	Supported	Supported
Transwarp Data Hub (TDH) 9.3.x	Supported	Supported	Supported	Supported	Supported	Supported
Transwarp ArgoDB	Supported	Supported	Not Supported	Supported	Supported	Supported
StarRocks	Supported	Supported	Not Supported	Supported	Supported	Supported
Lindorm (Compute Engine)	Supported	Supported	Not Supported	Supported	Supported	Supported
GaussDB (Data Warehouse Service)	Supported	Supported	Not Supported	Supported	Supported	Supported
Databricks	Supported	Supported	Not Supported	Supported	Supported	Supported
Amazon EMR	Supported	Supported	Not Supported	Supported	Supported	Supported
SelectDB	Supported	Supported	Not Supported	Supported	Supported	Supported
Doris	Supported	Supported	Not Supported	Supported	Supported	Supported
Aliyun EMR Serverless Spark	Supported	Supported	Not Supported	Supported	Supported	Supported

Assets

Engine type	Asset operations	Asset consumption
MaxCompute	Supported	Supported
AnalyticDB for PostgreSQL	Supported	Supported
E-MapReduce 3.x	Supported	Supported
E-MapReduce 5.x	Supported	Supported
CDH 5.x	Supported	Supported
CDH 6.x	Supported	Supported
Cloudera Data Platform 7.x	Supported	Supported
Huawei FusionInsight 8.x	Supported	Supported
AsiaInfo DP 5.3	Supported	Supported
Transwarp TDH 6.x	Supported	Supported
Transwarp TDH 9.3.x	Supported	Supported
Transwarp ArgoDB	Supported	Supported
StarRocks	Supported	Supported
Lindorm (Compute Engine)	Supported	Supported
GaussDB (DWS)	Supported	Supported
Databricks	Supported	Supported
Amazon EMR	Supported	Supported
SelectDB	Supported	Supported
Doris	Supported	Supported
Alibaba Cloud EMR Serverless Spark	Supported	Supported

Applicability

Engine type	Data service standard edition	Data operation enhancement	Tag platform offline edition	Tag platform real-time edition	Tag platform group selection
MaxCompute	Supported	Supported	Supported	Supported	Supported
AnalyticDB for PostgreSQL	Supported	Supported	Not Supported	Not Supported	Not Supported
E-MapReduce 3.x	Supported	Supported	Supported	Supported	Supported
E-MapReduce 5.x	Supported	Supported	Supported	Supported	Supported
Cloudera Distribution including Apache Hadoop (CDH) 5.x	Supported	Supported	Supported	Supported	Supported
Cloudera Distribution including Apache Hadoop (CDH) 6.x	Supported	Supported	Supported	Supported	Supported
Cloudera Data Platform 7.x	Supported	Supported	Supported	Supported	Supported
Huawei FusionInsight 8.x	Supported	Supported	Supported	Supported	Supported
AsiaInfo DP 5.3	Supported	Supported	Not Supported	Not Supported	Not Supported
Transwarp Data Hub (TDH) 6.x	Supported	Supported	Supported	Supported	Supported
Transwarp Data Hub (TDH) 9.3.x	Supported	Supported	Supported	Supported	Supported
Transwarp ArgoDB	Supported	Supported	Supported	Supported	Supported
StarRocks	Supported	Supported	Not Supported	Not Supported	Not Supported
Lindorm (Compute Engine)	Supported	Supported	Supported	Supported	Supported
GaussDB (Data Warehouse Service)	Supported	Supported	Not Supported	Not Supported	Not Supported
Databricks	Supported	Supported	Not Supported	Not Supported	Not Supported
Amazon EMR	Supported	Supported	Not Supported	Not Supported	Not Supported
SelectDB	Supported	Supported	Not Supported	Not Supported	Not Supported
Apache Doris	Supported	Supported	Not Supported	Not Supported	Not Supported
Alibaba Cloud EMR Serverless Spark	Supported	Supported	Not Supported	Not Supported	Not Supported

Platform

Engine type	Cross-tenant publishing	Row-level security	Query acceleration
MaxCompute	Supported	Supported	Supported
AnalyticDB for PostgreSQL	Supported	Supported	Not supported
E-MapReduce 3.x	Supported	Supported	Supported
E-MapReduce 5.x	Supported	Supported	Supported
CDH 5.x	Supported	Supported	Supported
CDH 6.x	Supported	Supported	Supported
Cloudera Data Platform 7.x	Supported	Supported	Supported
Huawei FusionInsight 8.x	Supported	Supported	Supported
AsiaInfo DP 5.3	Supported	Supported	Supported
Transwarp TDH 6.x	Supported	Supported	Not supported
Transwarp TDH 9.3.x	Supported	Supported	Not supported
Transwarp ArgoDB	Supported	Supported	Not supported
StarRocks	Supported	Supported	Not supported
Lindorm (Compute Engine)	Supported	Supported	Not supported
GaussDB (DWS)	Supported	Supported	Not supported
Databricks	Supported	Supported	Not supported
Amazon EMR	Supported	Supported	Not supported
SelectDB	Supported	Supported	Not supported
Doris	Supported	Supported	Not supported
Aliyun EMR Serverless Spark	Supported	Supported	Not supported

Resource usage

The number of Processing Unit Specifications and Quality Rules you can use depends on the Data Processing Units and Quotas you purchase. For more information, see View resource usage.

Create a tenant

Log in to the YuanCang Tenant as a Super Administrator or System Administrator.
In the top navigation bar of the Dataphin homepage, select Management Center > System Settings.
In the left-side navigation pane, select System Settings > Tenant Settings.
On the Tenant Settings page, click Add Tenant to open the Create Tenant dialog box.

In the Create Tenant dialog box, configure the following parameters.

Parameter		Description
Basic Information	Tenant Name	Enter a custom name for the tenant, up to 128 characters.
Basic Information	Super Administrator	Select a Super Administrator for the tenant. Note The Super Administrator cannot be changed after creation. Selecting Hide Super Administrator and System Administrator displays the role as "Administrator" in the UI but does not affect its underlying permissions.
Feature Settings	Engine Type	Select the Batch Compute Engine for the tenant. Different engines support different features. For more information, see Features supported by different engines. After you select an Engine Type, configure the available features for the tenant below. Data Processing Unit: Specifies the number of Data Processing Units. This value must be greater than 0 and not exceed the available quota. Rule Quota: This quota is shared between Domain Quality Rules and Global Quality Rules. Only enabled quality rules count toward this quota. The upper-left corner of the tenant list displays the totals for purchased and allocated Data Processing Units and Quality Rules. Note The Engine Type selected here does not restrict the engine types users can select within the tenant. For multi-engine tenants, you cannot configure the Engine Type. The feature section defaults to all supported features. By default, each tenant can create one consumption channel. You can purchase additional channels. You can create Data Source monitoring rules if you enable either Domain Quality or Global Quality. If you purchase both, these rules are counted as Domain Quality Rules. If you only enable Global Quality, they are counted as Global Quality Rules. To use X-Data Engineering, you must enable the Intelligent Development Edition. To use X-Analysis, X-Asset Q&A, and X-Catalog Management, you must enable Asset Operation The Data Operation Enhancement feature requires the Data Service Standard Edition.
Scheduling Resource Settings	CPU/Memory Allocatable Resources	Note This section allocates Scheduling Resources only for the default Dataphin cluster. The Minimum Available value represents the guaranteed resources for the tenant. If total resource demand exceeds the supply, each tenant is guaranteed its configured minimum amount. Other tenants can borrow the unused portion of a tenant's minimum available resources. However, the original tenant can reclaim these resources through a queuing mechanism when needed. A tenant can also borrow unused resources from other tenants, but the total resources used cannot exceed its configured Maximum Available limit. Enter the Minimum Available and Maximum Available values for CPU and Memory. Minimum Available: A value greater than or equal to 0, with one decimal place. The value cannot exceed the remaining allocatable resources. The sum of the Minimum Available values for all tenants cannot exceed 100%. Maximum Available: A value greater than or equal to 0, with up to one decimal place. The value must be greater than or equal to the Minimum Available value and less than 100%.
Scheduling Resource Settings	Metadata Processing Resource Group	A metadata collection task is split into multiple sub-tasks during execution: Fetching initial metadata from a Data Source for project initialization uses the collection task's specified scheduling Resource Group to ensure network connectivity. Subsequent processing and data backflow tasks use the Resource Group configured here to reduce the impact on other tasks. We recommend using a Resource Group from the Dataphin cluster. Note If the CPU or Memory for Scheduling Resources is configured to a value other than 0, you can select the Default resource group of the current tenant. Changes to the Resource Group configuration take approximately 10 minutes to apply. Instances generated before the change are not affected.
Cross-tenant Settings	Cross-tenant Dependency	If enabled, tasks in this tenant can depend on tasks in other tenants, and vice versa.

Click OK to create the tenant.

Manage tenants

After you create a tenant, you can edit, delete, or restore it on the Tenant Settings page.

Section

Description

① Tenant List

Displays the Tenant Name/ID, Super Admin, Used Quota and Total Quota for Data Processing Units and Quality Rules, Resource Scope Settings, Currently Used Resources, and Feature Modules.

Resource Scope Settings: The configured minimum and maximum usage limits for Scheduling Resources. This does not represent actual usage.
Currently Used Resources: The amount of resources allocated to the tenant, as of the last page refresh. This represents the tenant's actual resource usage.

② Actions

You can Edit, Delete, and Restore tenants.

You cannot perform any actions on a tenant with the Creating status. You also cannot edit or delete a tenant with the Updating status. If the status is Creation Failed or Update Failed, you can edit the tenant again. Editing a tenant that failed to update re-initializes it.

Note

The system reserves allocated Data Processing Units, Scheduling Resources, and Quality Rule resources for a tenant in the Creating status. If creation fails, these resources are released.

Edit: You can edit the tenant name, Data Processing Unit Quota, and Feature Modules. The engine type cannot be edited. For a Metadata Warehouse Tenant, you can set the minimum and maximum usage limits for CPU and memory.
Feature downgrade has the following limitations:
- You cannot disable the Intelligent Development feature if the tenant contains related objects, such as Business Segments, Business Objects, Data Domains, Business Activities, Logical Tables, Business Qualifiers, Atomic Metrics, or Derived Metrics. To disable the feature, you must first delete these objects.
- You cannot disable a Feature Module if it has existing configurations. To disable the feature, you must first delete the corresponding configurations. For example:
  If Real-time Integration Tasks exist (including tasks in Draft, Development, or Production states), you cannot disable Real-time Integration.
  If baselines or throttling rules exist, you cannot disable Intelligent O&M.
- You cannot disable the Basic Development Edition.
Note
If the custom Resource Group feature is enabled for your account, you can modify the amount of Scheduling Resources allocated to the Metadata Warehouse Tenant. Allocating insufficient resources to the Metadata Warehouse Tenant may cause delays in system task completion. This can affect the display of some features, such as field popularity on the catalog page, and delay data updates in the metadata sharing model.
Delete: To delete a tenant, find it in the list and click the Delete icon in the Actions column. In the Delete Tenant dialog, enter the 6-digit Verification Code and click OK.
Note
- You cannot delete the Metadata Warehouse Tenant.
- When you delete a tenant, it is soft-deleted and remains in the tenant list for 30 days. You can restore the tenant during this period. After 30 days, the tenant is permanently removed from the list. The tenant's data is not deleted automatically. To permanently delete the data, contact Technical Support.
- To delete a tenant that has custom Resource Groups, you must first delete those Resource Groups.
Restore: To restore a deleted tenant, select it from the list. In the Restore Tenant dialog, click OK.
If you cannot restore a tenant, it may be for one of the following reasons:
- The number of Data Processing Units used by the tenant exceeds the remaining allocatable quantity.
- The number of Quality Rules used by the tenant exceeds the remaining allocatable quantity.
- One or more features the tenant used have been downgraded since its deletion.
- The tenant's configured Scheduling Resources (CPU or memory) exceed the system's remaining allocatable capacity.