All Products
Search

This Product

    PolarDB:Temporarily add compute nodes

    Document Center

    PolarDB:Temporarily add compute nodes

    Last Updated:Mar 30, 2026

    When your cluster faces predictable short-term spikes in read traffic—such as e-commerce promotions, batch analytics, or report generation—temporarily adding read-only nodes lets you scale read capacity on demand without permanently changing your cluster configuration. Set a restore point, and the nodes are automatically removed when the time is reached, with no manual intervention required.

    Prerequisites

    Before you begin, make sure that:

    • The cluster uses Subscription billing

    • The cluster is in the Running state, with no configuration change tasks (such as adding nodes, deleting nodes, or minor version upgrades) in progress

    • There are no pending configuration change tasks, such as adding nodes, deleting nodes, or changing configurations

    • The cluster has at least one read-only node—you cannot add temporary nodes to a single-node cluster

    • To add temporary nodes with a columnstore index (In-Memory Columnar Index, or IMCI): the cluster has at least one read-only node with a columnstore index enabled

    • The cluster is not a Multi-master Cluster (Limitless) Edition cluster—that edition does not support temporary read-only nodes

    Limitations

    Area Limitation
    Service impact — adding nodes Adding a temporary read-only node takes approximately 5 minutes. Multiple nodes added at the same time are processed in parallel. The cluster is not affected during the addition.
    Service impact — removing nodes When a node is automatically removed at the restore point, connections to that node experience a brief disconnection. Other nodes are not affected. If your application connects through the cluster endpoint, PolarDB automatically blocks the deleted node—no application changes needed.
    Node quantity — Enterprise Edition (Cluster Edition) Up to 15 read-only nodes total. At least one read-only node must remain at all times.
    Node quantity — Multi-master Cluster (Limitless) Edition Up to 63 read/write nodes and 15 global read-only nodes. Temporary read-only nodes are not supported.
    Node quantity — Standard Edition Up to 7 read-only nodes total.
    Node specification — memory A new read-only node's memory cannot be less than half of the primary node's memory.
    Node specification — CPU cores Minimum CPU core requirements based on the primary node (see table below). Not applicable to Multi-master Cluster (Limitless) Edition.
    Concurrent operations During a temporary node addition period, you can perform this operation only one more time.
    Blocked operations While temporary nodes exist, you cannot delete nodes, manually upgrade or downgrade the cluster, perform a temporary upgrade, or terminate the cluster.
    Duration 1 hour to 14 days. For longer durations, use manual upgrade or downgrade instead.
    Restore point Cannot be later than one day before the cluster's expiration date. For example, if the cluster expires on January 10, the latest restore point is January 9. Once set, the restore point cannot be modified.
    Extending the duration Add another temporary node before the restore time. The new restore point cannot be earlier than the previously set time—the duration must extend by at least one hour.

    Minimum CPU core requirements for read-only nodes:

    Primary node CPU cores Minimum read-only node CPU cores
    2 2
    4 4
    8 4
    16 8
    32 16
    64 32
    88 64
    120 64

    Add temporary read-only nodes

    1. Log in to the PolarDB console. In the left navigation pane, click Clusters and select the region where the cluster is located.

    2. Open the Add/Remove Node dialog using either of the following methods:

      • In the cluster list, click Add/Remove Node in the Actions column for the target cluster.

      • Click the cluster ID to go to the Basic Information page. In the Database Nodes section, click Add/Remove Node.

    3. In the Add/Remove Node dialog, select one of the following options and click OK:

      • Expenses and Costs > Orders > My OrdersTemporarily Add Node: adds a standard read-only node. Use this for general read traffic scaling.

      • Temporarily Add IMCI Read-only Nodes: adds a read-only node with a columnstore index. Use this to accelerate analytical or reporting queries that benefit from columnar storage.

    4. On the Temporary Upgrade page, review the Current Configuration and Expiration Time. Click Add Node or Add a Read-only IMCI Node and select the node specification.

      Click the button multiple times to add more than one node at the same time.

    5. Set a Restore Point—the date and time at which the system automatically removes the temporary nodes and reverts to the original configuration.

    6. Review the service agreements, click Buy Now, and complete the payment. The temporary node addition starts after payment is confirmed.

    Billing

    The cost is calculated based on the price difference between the new and original specifications, with a 1.5x premium coefficient:

    Cost per compute node for N days = (Monthly subscription price of the specification / 30) × 1.5 × N

    Manage temporary nodes

    Convert temporary nodes to permanent nodes

    If you want to keep the temporary nodes permanently—or if you need to perform operations that are blocked while temporary nodes exist (such as deleting nodes, upgrading or downgrading, or terminating the cluster)—convert the temporary nodes to permanent nodes first. Permanent nodes are not automatically removed at the restore point.

    Manual conversion: During the temporary node addition period, open the Add/Remove Node wizard again and run the Add Nodes operation. This creates a permanent billing order without causing a service interruption.

    Conversion triggered by storage scale-out: If you manually scale out the cluster's storage capacity while temporary nodes exist, the system automatically converts the temporary nodes to permanent nodes. Billing switches to the subscription price of the new specifications from the time of the scale-out. Evaluate this carefully before scaling out storage.

    View the expiration time of temporary nodes

    Go to [Expenses and Costs](https://usercenter2-intl.console.alibabacloud.com/billing/#/account/overview) > Orders > My Orders. Search by cluster ID, find the order with an order type of Temporary Upgrade, and click Details.

    API reference

    API Description
    TempModifyDBNode Temporarily adds nodes to a PolarDB cluster

    FAQ

    What do I do if I get an "insufficient resources" error?

    The selected specification is sold out in the current zone. Try a similar specification with slightly fewer resources. If the issue recurs, change the primary zone to avoid running into the same problem during future scaling operations.

    What do I do if I get a "no specification change" prompt?

    No nodes have been added. Click Temporarily Add Node or Temporarily Add IMCI Read-only Nodes to select and add a read-only node.

    How do I delete nodes or change node configurations when temporary nodes exist?

    Convert the temporary nodes to permanent nodes first. After conversion, you can delete nodes or change configurations as usual.

    What does "Temporary upgrade expiration time cannot be earlier than: XXX" mean?

    After you add temporary nodes, you can extend the duration by adding another round of temporary nodes before the scheduled restore time. The new restore point must be later than the current one—the duration must extend by at least one hour.

    My cluster has high write pressure. Will adding read-only nodes help?

    No. Read-only nodes handle read (SELECT) requests only. All write operations (INSERT, UPDATE, DELETE) are processed by the primary node. To address write pressure, upgrade the primary node's specification instead.

    References