You can scale the computing resources and storage capacity of an AnalyticDB for MySQL cluster to meet your business requirements. You can also change the mode of a cluster.

Prerequisites

No renewal orders are overdue within your Alibaba Cloud account.

Background information

AnalyticDB for MySQL supports the following configuration change operations:
  • Scale up or down a cluster: For clusters in Reserved Mode, you can change the cluster specifications, number of node groups, and storage capacity. For clusters in Elastic Mode, you can change the computing resources and elastic I/O resources.
  • Change the cluster mode: You can change the cluster mode from Reserved Mode to Elastic Mode.

Precautions

  • Scale up or down a cluster
    • During a scaling event, you are not allowed to execute the SUBMIT JOB statement to submit asynchronous tasks. If your business requires asynchronous tasks, perform scaling during appropriate periods.
    • When a cluster is scaled up or down, data in the cluster is migrated for redistribution. The amount of time required for data migration is proportional to the volume of data.
    • During a scale-down event, data migration can take up to dozens of hours to complete. Proceed with caution especially when your cluster contains a large amount of data.
    • When a scaling event is about to end, you may encounter transient connections. We recommend that you scale your cluster during off-peak hours or make sure that your application is configured to automatically reconnect to your cluster.
    • If you change the enhanced SSD (ESSD) performance level, you cannot change other configurations.
  • Change the cluster mode
    • The cluster mode can be changed only for AnalyticDB for MySQL clusters of minor engine version 3.1.3.9 or later.

      For more information about how to view the minor engine version of a cluster, see How can I view the version of an AnalyticDB for MySQL cluster? To update the minor engine version of a cluster, Submit a ticket.

    • When you change the cluster mode, you must apply for new cluster resources, perform full and incremental migration, and switch your business from the existing cluster to the new cluster. The amount of time it takes to complete these tasks depends on the cluster specifications and data volume.
    • During the switchover from the existing cluster to the new cluster, the existing cluster becomes read-only for a period of 5 to 10 minutes.
      After you connect to the existing cluster, execute the following statement to specify the time window for the switchover. In this example, the switchover is performed within a time window of 23:00 to 23:30.
      SET adb_config REPLICATION_SWITCH_TIME_RANGE=`23:00, 23:30`;
      For more information about how to connect to a cluster, see Connect to an AnalyticDB for MySQL cluster.

Billing

For information about the billing of configuration changes, see Billing of configuration changes.

Scale up a cluster

Note
  • For a cluster in Reserved Mode, you can change the cluster mode to Elastic Mode or increase one or all of the cluster specifications, number of node groups, and storage capacity.
  • For a cluster in Elastic Mode, you can increase one or both of the computing resources and elastic I/O resources.
  1. Log on to the AnalyticDB for MySQL console.
  2. In the upper-left corner of the page, select the region where the cluster resides.
  3. In the left-side navigation pane, click Clusters.
  4. On the Data Warehouse Edition (V3.0) tab, find the cluster that you want to scale up and click Scale Up in the Actions column.
  5. On the Upgrade/Downgrade page, configure the parameters that are described in the following table.
    Table 1. Parameters
    ParameterDescription
    VersionThe version of the cluster. This parameter is automatically set to 3.0. You do not need to specify this parameter.
    ModeThe mode of the cluster. Valid values: Elastic Mode and Reserved Mode.
    Note You can change the cluster mode from Reserved Mode to Elastic Mode but not from Elastic Mode to Reserved Mode. For more information about the differences between the two modes, see FAQ.
    EditionThe edition of the cluster. The value of the Edition parameter depends on that of the Mode parameter. The following section describes the editions supported by different modes:
    • When the Mode parameter is set to Elastic Mode, Elastic Cluster is automatically selected. You do not need to specify this parameter.
    • When the Mode parameter is set to Reserved Mode, Reserved Cluster is automatically selected. You do not need to specify this parameter.
    Note If you change the cluster mode from Reserved Mode to Elastic Mode and you need to use external tables, you must turn on ENI in the Network Information section of the Cluster Information page before you can access data of external tables. When you turn on ENI, cluster performance and security are not affected.
    Instance TypeThe instance type of the cluster.
    • If you set the Mode parameter to Elastic Mode and the Edition parameter to Elastic Cluster, the instance type is automatically set to E32. You do not need to specify this parameter.
    • If you set the Mode parameter to Reserved Mode and the Edition parameter to Reserved Cluster, the instance type can be set to C8 or C32.
    Computing ResourcesThe computing resources of the cluster. Computing resources are used for data computing. More computing resources can speed up data queries. We recommend that you select more CPU cores for computing resources than those for storage resources.
    Note This parameter is available only if the Mode parameter is set to Elastic Mode.
    ESSD Performance LevelThe ESSD performance level of the cluster. ESSD performance levels provide the following capacities:
    • PL1: 20 GiB to 32,768 GiB
    • PL2: 461 GiB to 32,768 GiB
    • PL3: 1,261 GiB to 32,768 GiB
    Important
    • If you change the ESSD performance level, you cannot change other configurations.
    • If you want to change the ESSD performance level from PL1 to PL2 or from PL2 to PL3, the amount of hot data must reach the minimum capacity of the selected performance level.
    • This configuration changes the ESSD performance level but does not purchase storage. You are charged for the storage that is actually occupied.
    EIU CapacityThe single-node specifications of an elastic I/O unit (EIU) for the cluster. Each EIU consists of three nodes.
    • If you select 8 cores and 64 GB, the specifications of an EIU are 24 cores and 192 GB memory.
    • If you select 12 cores and 96 GB, the specifications of an EIU are 36 cores and 288 GB memory.

    Elastic I/O resources are bundled into units and sold as EIUs. You can use EIUs to individually scale up storage resources for clusters in Elastic mode for Cluster Edition. For more information, see Use EIUs to scale up storage resources.

    Note You can specify this parameter only if the Mode parameter is set to Elastic Mode and the Computing Resources parameter is set to 32 Cores and 128 GB or more.
    EIUsThe number of EIUs.
    The number of EIUs that you can purchase varies based on the single-node EIU specifications.
    • If the single-node EIU specifications are 8 cores and 64 GB, you can purchase up to 32 EIUs.
    • If the single-node EIU specifications are 12 cores and 96 GB, you can purchase up to 16 EIUs.
    Note You can specify this parameter only if the Mode parameter is set to Elastic Mode and the Computing Resources parameter is set to 32 Cores and 128 GB or more.
    Node GroupsThe number of node groups. By default, each node group consists of three nodes.
    Note This parameter is available only if the Mode parameter is set to Reserved Mode.
    CapacityThe storage capacity of a single node group. If you purchase multiple node groups, you can use the following formula to calculate the total storage capacity of the cluster: Total storage capacity = Storage capacity per node group × Number of node groups.
    Note This parameter is available only if the Mode parameter is set to Reserved Mode.
  6. Complete the subsequent operations based on the Product Type value of the cluster before the cluster is scaled up.
    • If Product Type is set to Subscription before the cluster is scaled up, perform the following operations:
      1. Read and select the Terms of Service, and then click Buy Now.
      2. On the Purchase page, confirm the order and the payment method, and then click Purchase.
    • If Product Type is set to Pay-As-You-Go before the cluster is scaled up, perform the following operations:

      Read and select the Terms of Service, and then click Buy Now.

Scale down a cluster

  1. Log on to the AnalyticDB for MySQL console.
  2. In the upper-left corner of the page, select the region where the cluster resides.
  3. In the left-side navigation pane, click Clusters.
  4. On the Data Warehouse Edition (V3.0) tab, find the cluster that you want to scale down and choose More > Scale Down in the Actions column.
  5. On the Downgrade page, configure parameters of the cluster. For more information about the parameters, see the "Parameters" section of this topic.
  6. Complete the subsequent operations based on the Product Type value of the cluster before the cluster is scaled down.
    • If Product Type is set to Subscription before the cluster is scaled down, perform the following operations:
      1. Read and select the Terms of Service, and then click Buy Now.
      2. On the Purchase page, confirm the order and the payment method, and then click Purchase.
    • If Product Type is set to Pay-As-You-Go before the cluster is scaled down, perform the following operations:

      Read and select the Terms of Service, and then click Buy Now.