Create and manage resource groups

1. Log on to the AnalyticDB for MySQL console. In the upper-left corner of the console, select a region. In the left-side navigation pane, click Clusters. Find the cluster that you want to manage and click the cluster ID.

2. In the left-side navigation pane, choose Cluster Management > Resource Management and click the Resource Groups tab. In the upper-right corner of the resource group list, click Create Resource Group.

3. Enter a name for the resource group and select a Job Type.

   • Select Interactive for online workloads that require high QPS and low latency.

     An Interactive resource group uses resident computing resources and runs queries in MPP mode to deliver responses in milliseconds.

   • Select Job for offline workloads that require high throughput.

     A Job resource group provisions temporary computing resources to run queries in BSP mode, with response times in seconds or minutes. The amount of resources provisioned ranges from 0 ACU to the group's maximum limit, depending on the running task's size.

   • Select AI for heterogeneous computing workloads.

     An AI resource group supports heterogeneous computing with GPUs and CPUs and various deployment types such as MLSQL models and Ray-managed computing.

   Important

   You cannot change the task type after the resource group is created.

4. The configurable properties depend on the task type you select. After setting the properties, click OK.

   Interactive resource group properties

   Parameter	Description
   Engine	XIHE: The resource group can only execute XIHE SQL queries. Spark: The resource group can only execute Spark SQL jobs, which are processed in interactive mode. Important You cannot change the engine after the resource group is created.
   Automatic Stop	If an Interactive resource group is idle for a specified period, its running clusters are automatically released. This feature reduces costs by preventing resource waste, but restarting resources for the next query introduces a delay. Important This parameter is available only when the Engine is set to Spark.
   Cluster Size	If the Engine is XIHE: The size of a single cluster in ACUs. The minimum value is 16 ACU. If the Engine is Spark: The size of a single cluster, which is the number of ACUs allocated to a Spark application. The minimum value is 24 ACU. Each Spark Interactive resource group can run multiple Spark applications. The Minimum Clusters and Maximum Clusters settings determine the number of Spark applications that can run in the resource group. For information about the mapping between cluster size and Spark Driver/Executor specifications, see Appendix: Mapping between cluster size and Spark Driver and Spark Executor specifications.
   Minimum Clusters Maximum Clusters	Minimum Clusters: The minimum number of clusters that must run in the resource group. The minimum value is 1. Maximum Clusters: The maximum number of clusters that the resource group can scale out to. The maximum value is 10. If Minimum Clusters and Maximum Clusters are different, AnalyticDB for MySQL dynamically scales the number of clusters within this range based on query load. If Minimum Clusters and Maximum Clusters are the same, AnalyticDB for MySQL starts a fixed number of clusters, giving you static control over the group's total computing resources. Note If Minimum Clusters or Maximum Clusters is 2 or greater, the Multi-Cluster feature is enabled for the resource group. For more information, see Multi-Cluster elastic model.
   Job Resubmission Rules	Routes queries that exceed the Query Execution Time Threshold to the specified Target Resource Group. For more information, see Job resubmission. Important This parameter is available only when the Engine is set to XIHE.
   Spark Configuration	Sets configuration parameters for all Spark jobs in this resource group. To configure a specific job, set its parameters in the submission code. For more information about Spark configuration parameters, see Spark application configuration parameters. Important This parameter is available only when the Engine is set to Spark.

   Job resource group properties

   Parameter	Description
   Minimum Computing Resources	The minimum value is 0 ACU. Important The minimum computing resources cannot be changed after the resource group is created.
   Maximum Computing Resources	In the console, you can set the maximum computing resources up to 1,024 ACU in increments of 8 ACU. To request a higher limit, you can submit a ticket.
   Spot Instance	Specifies whether to enable spot instances. If you enable this option, Spark jobs that run in the Job resource group attempt to use spot instance resources. For more information, see Spot instances.
   Spark Configuration	Sets configuration parameters for all Spark jobs in this resource group. To configure a specific job, set its parameters in the submission code. For more information about Spark configuration parameters, see Spark application configuration parameters.

   AI resource group

   Parameter	Description
   Deployment Mode	Select RayCluster.
   Head Node Resource Type	Supports CPU and GPU types. CPU: Use for routine computing tasks, multitasking, or complex logical operations. We recommend using the CPU type for production environments. GPU: Use for large-scale parallel data processing, machine learning, or deep learning training. In development and test environments, you can select the GPU type if you need GPU resources without creating worker nodes.
   Head Resource Specifications	If the Head Resource Specifications is CPU, you can select specifications such as small, m.xlarge, and m.2xlarge. The CPU core count for each specification is the same as for Spark resource specifications. For more information, see Spark resource specification list. If the Head Resource Specifications is GPU, contact technical support for assistance with model selection due to GPU model and inventory constraints. Important The head node primarily handles job scheduling. Select a specification based on the overall scale of your Ray Cluster.
   Head disk space	Used to store Ray logs, temporary data, and spilled object data. Unit: GB. Range: 30 to 2,000. Default: 100. Important This disk is for temporary storage only. Do not use it for long-term data.
   Allocation Unit	The number of GPUs to allocate per head node. For example, a value of 1/3 assigns one-third of a GPU to each head node. Important This parameter is required only when the Head Resource Specifications is GPU.
   Worker Group Name	A custom name for the Worker Group. An AI resource group can have multiple Worker Groups with different names.
   Worker Resource Type	Supports CPU and GPU types. If your workload involves routine computing tasks, multitasking, or complex logical operations, we recommend CPU. If your workload involves large-scale parallel data processing, machine learning, or deep learning training, we recommend GPU.
   Worker Resource Specifications	If the Worker Resource Type is CPU, you can select specifications such as small, m.xlarge, and m.2xlarge. The CPU core count for each specification is the same as for Spark resource specifications. For more information, see Spark resource specification list. If the Worker Resource Type is GPU, submit a ticket to contact technical support for assistance with model selection due to GPU model and inventory constraints.
   Worker Disk Storage	Used to store Ray logs, temporary data, and spilled object data. Unit: GB. Range: 30 to 2,000. Default: 100. Important This disk is for temporary storage only. Do not use it for long-term data.
   Minimum Workers Maximum Workers	Minimum Workers: The minimum number of workers that must run in a Worker Group. The minimum value is 0. Maximum Workers: The maximum number of workers that can run in a Worker Group. The maximum value is 8. Worker Groups support auto-scaling, and each Worker Group can scale independently. If the minimum and maximum worker counts differ, AnalyticDB for MySQL scales the number of workers within this range based on task load. For clusters with multiple Worker Groups, the system automatically selects the best-fit group for a task, preventing individual groups from being overloaded or idle.
   Allocation Unit	The number of GPUs to allocate per worker node. For example, a value of 1/3 assigns one-third of a GPU to each worker node. Important This parameter is required only when the Worker Resource Type is GPU.

Modifiable properties

• For custom resource groups (resource groups that you create), you can modify the following properties:

  • For Interactive resource groups: Auto Stop, Cluster size, Minimum clusters, Maximum clusters, Job resubmission rule, and Spark configuration.

  • For Job resource groups: Maximum computing resources, Spot instance, and Spark configuration.

  • For AI resource groups (Ray Cluster deployment type): Head resource specifications, Worker resource type, Worker resource specifications, Worker disk space, Minimum workers, and Maximum workers.

  Other properties cannot be modified, including Resource Group Name, Task type, the Engine for an Interactive resource group, the Minimum computing resources for a Job resource group, and the Deployment Type and Worker Group Name for an AI resource group.

• For default resource groups (named user_default and serverless):

  • For Enterprise Edition and Basic Edition, you can only modify the job resubmission rule for the user_default resource group. The serverless resource group cannot be modified.

  • For Data Lakehouse Edition, you can modify the Reserved Computing Resources and job resubmission rule for the user_default resource group. The serverless resource group cannot be modified.

Procedure

1. On the Resource Groups page, find the target resource group and click Modify in the Actions column.

2. In the Modify Resource Group panel, modify the properties and click OK.

   The changes take effect when the resource group's status becomes "Running".

Modifiable properties

After a resource group is created, you can modify its query type or resource amount.

• For the default resource group (named user_default), you can only modify the query type. You cannot manually modify the resource amount.

  Note

  The resource amount for the default resource group is calculated as: Total cluster resources minus the resources used by all other resource groups in the cluster.

• For custom resource groups (resource groups that you create), you can modify both the query type and the resource amount.

Procedure

1. On the Resource Groups page, find the target resource group and click Modify in the Actions column.

2. Modify the Query Type or Resource Amount as needed.

3. After you are finished, click OK.

   Changes to the resource amount of an AnalyticDB for MySQL resource group take effect immediately.