AnalyticDB:Create, modify, and delete a resource group

Prerequisites

An AnalyticDB for MySQL Enterprise Edition, Basic Edition, or Data Lakehouse Edition cluster is created.

Billing rules

• You are charged fees for the AnalyticDB compute unit (ACU) elastic resources used by interactive resource groups and job resource groups.

• When you create an AI resource group in RayCluster deployment mode, you are charged for the corresponding resources.

  • If you set the Worker Resource Type parameter to CPU, you are charged for the ACU elastic resources used by the AI resource group.

  • If you set the Worker Resource Type parameter to GPU, you are charged based on the GPU specifications and quantity of the AI resource group.

  • You are charged based on the storage size specified by the Worker Disk Storage parameter.

You can use the following methods to view the amount of elastic resources used by the resource group:

• Enterprise Edition and Basic Edition: In the left-side navigation pane, choose Cluster Management > Resource Management. On the Resource Overview tab, you can view the total amount of resources and that of reserved resources used by all resource groups. You can use the following formula to calculate the amount of elastic resources: Total amount of resources - Amount of reserved resources.

• Data Lakehouse Edition: In the left-side navigation pane, choose Cluster Management > Resource Management. On the Resource Overview tab, you can view the total amount of computing resources and that of reserved computing resources used by all resource groups. You can use the following formula to calculate the amount of elastic resources: Total amount of computing resources - Amount of reserved computing resources.

Create a resource group

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. On the Resource Management page, click the Resource Groups tab. In the upper-right corner of the Resource Groups tab, click Create Resource Group.

3. In the Create Resource Group panel, enter a resource group name and select an option from the Job Type drop-down list.

   • If you want to perform real-time analysis with high queries per second (QPS) and low response time (RT), select Interactive.

     An interactive resource group uses resident computing resources to execute queries in massively parallel processing (MPP) mode. In this mode, query requests are responded within milliseconds.

   • If you want to perform batch processing with high throughput, select Job.

     A job resource group pulls up temporary computing resources to execute queries in bulk synchronous parallel (BSP) mode. In this mode, query requests are responded within seconds or minutes. The amount of temporary computing resources that can be pulled up varies based on the size of the job that you run, ranging from 0 ACUs to the maximum resource amount of the job resource group.

   • If you want to perform heterogeneous computing, select AI.

     An AI resource group uses GPU and CPU heterogeneous computing resources to support various deployment modes such as MLSQL and RayCluster.

   Important

   After you create a resource group, you cannot change the job type of the resource group.

4. Configure the property parameters of the resource group based on the selected job type. Then, click OK.

   Properties of an interactive resource group

   Parameter	Description
   Engine	XIHE: The resource group supports only XIHE SQL jobs. Spark: The resource group supports only Spark SQL jobs. Spark SQL jobs are run in an interactive manner. Important After you create a resource group, you cannot change the engine of the resource group.
   Automatic Stop	Specifies whether to enable the automatic stop feature for the resource group. If an interactive resource group remains idle for a specific period of time after all SQL statements are executed, the system uses the automatic stop feature to automatically release the compute clusters that are started in the resource group. The automatic stop feature prevents resource waste and reduces resource costs. However, if you execute queries after the started compute clusters are released, resources need to be re-pulled up and latency occurs. Important This parameter is available only if you set the Engine parameter to Spark.
   Cluster Size	The size of a single compute cluster when the Engine parameter is set to XIHE. Unit: AnalyticDB compute units (ACUs). The minimum value is 16 ACUs. The size of a single compute cluster when the Engine parameter is set to Spark. Set this parameter to the number of ACUs that you want to allocate to each Spark application. The minimum value is 24 ACUs. You can run multiple Spark applications in each Spark interactive resource group. You can specify the Minimum Clusters and Maximum Clusters parameters to define the range of Spark applications that can run in the resource group. For more information, see the "Mappings between the compute cluster size and Spark driver and executor specifications" section of the Multi-cluster scaling models topic.
   Minimum Clusters Maximum Clusters	Minimum Clusters: the minimum number of compute clusters that you must run in the resource group. The minimum value is 1. Maximum Clusters: the maximum number of compute clusters that you can run in the resource group. The maximum value is 10. If you specify different values for the Minimum Clusters and Maximum Clusters parameters, AnalyticDB for MySQL dynamically scales the number of compute clusters available for the resource group within the range of the minimum and maximum values based on query loads. If you specify the same value for the Minimum Clusters and Maximum Clusters parameters, AnalyticDB for MySQL starts the specified compute clusters after the resource group is created to statically maintain the total amount of computing resources available for the resource group. Note If you set the Minimum Clusters or Maximum Clusters parameter to a value that is greater than or equal to 2, the multi-cluster feature is enabled for the resource group. For information about the multi-cluster feature, see Multi-cluster scaling models.
   Job Resubmission Rules	The rules that resubmit the queries executed for a period of time longer than the specified execution duration threshold to the destination resource group. For more information, see Job resubmission. Important This parameter is available only if you set the Engine parameter to XIHE.
   Spark Configuration	The Spark application configuration parameters that can be applied to all Spark jobs executed in the resource group. If you want to configure parameters for a specific Spark job, you can specify values for the parameters in the code editor when you submit the job. For information about the Spark configuration parameters, see Spark application configuration parameters. Important This parameter is available only if you set the Engine parameter to Spark.

   Properties of a job resource group

   Parameter	Description
   Minimum Computing Resources	The minimum amount of computing resources. Set the value to 0 ACUs. Important After you create a job resource group, you cannot change the minimum computing resources.
   Maximum Computing Resources	The maximum amount of computing resources. Set this parameter within the range of 8 ACUs to 1,024 ACUs in increments of 8 ACUs. If you want to change the maximum value of the Maximum Computing Resources parameter, submit a ticket.
   Spot Instance	Specifies whether to enable the spot instance feature for the job resource group. After you enable the spot instance feature for the job resource group, Spark jobs that run in the job resource group attempt to use the spot instance resources. For more information, see Spot instances.
   Spark Configuration	The Spark application configuration parameters that can be applied to all Spark jobs executed in the resource group. If you want to configure parameters for a specific Spark job, you can specify values for the parameters in the code editor when you submit the job. For information about the Spark configuration parameters, see Spark application configuration parameters.

   Properties of an AI resource group

   Parameter	Description
   Deployment Mode	The deployment mode of the resource group. Select RayCluster.
   Head Resource Specifications	The head node is responsible for managing Ray metadata, running the Global Control Store (GCS) service, and scheduling tasks, but does not execute tasks. The head resource specifications refer to the number of CPU cores. You can choose specifications such as small, m.xlarge, and m.2xlarge. The number of CPU cores is the same between head resource specifications and Spark resource specifications. For more information, see Spark resource specifications. Important The head node is responsible for job scheduling. Select the head resource specifications based on the overall scale of the Ray cluster.
   Worker Group Name	The name of the worker group. You can configure multiple worker groups with different names in one AI resource group.
   Worker Resource Type	The type of the worker group, which can be CPU or GPU. If your business involves daily computing tasks, multitasking, or complex logical operations, we recommend that you select CPU. If your business involves large-scale data parallel processing, machine learning, or deep learning training, we recommend that you select GPU.
   Worker Resource Specifications	If you set the Worker Resource Type parameter to CPU, you can select specifications such as small, m.xlarge, and m.2xlarge. The number of CPU cores is the same between head resource specifications and Spark resource specifications. For more information, see Spark resource specifications. If you set the Worker Resource Type parameter to GPU, submit a ticket for technical assistance because the specifications are related to GPU models and inventory.
   Worker Disk Storage	The disk storage is used to store Ray logs, temporary data, and overflow data from Ray distributed object storage. Unit: GB. Valid values: 30 to 2000. Default value: 100. Important Disks are used to temporarily store data. Do not use disks for long-term data storage.
   Minimum Workers Maximum Workers	Minimum Workers: the minimum number of worker nodes that are required in a worker group, with a minimum value of 1. Maximum Workers: the maximum number of worker nodes that are allowed in a worker group, with a maximum value of 8. Each worker group can be automatically scaled. If the minimum number of worker nodes is different from the number of maximum worker nodes in a worker group, AnalyticDB for MySQL dynamically changes the number of worker nodes within the range based on the number of tasks. If multiple worker groups exist, AnalyticDB for MySQL performs automatic matching to prevent overloading or idling of a single worker group.
   Distribution Unit	The number of GPUs that are allocated to each worker node. Example: 1/3. Important This parameter is required only when you set the Worker Resource Type parameter to GPU.

Modify a resource group

Delete a resource group

You can delete custom resource groups, but you cannot delete the default resource groups named user_default and serverless.

Monitor the resource usage

You can view the resource usage at the cluster level, the resources and workloads at the resource group level, and the resource consumption at the job level. For more information about the resource usage metrics, see the "Resource group monitoring" section of the Metric overview topic.

Associate with or disassociate from a database account

If you want to isolate resources among queries that are initiated by different database accounts, you can associate different resource groups with the database accounts. After you associate a resource group with a database account, the system routes the SQL queries to the associated resource group if you do not specify a resource group for SQL queries initiated by the database account.

FAQ

References

• For information about the functionalities of resource groups and how to specify a resource group to execute queries, see Resource group management (Data Lakehouse Edition).

• For information about how to create, modify, and delete resource groups and associate resource groups with database accounts by calling API operations, see Resource group management.