All Products
Search

This Product

    E-MapReduce:Scale out a cluster

    Document Center

    E-MapReduce:Scale out a cluster

    Last Updated:Aug 22, 2026

    Scale out an E-MapReduce cluster by adding core or task nodes when compute or storage resources are insufficient.

    Prerequisites

    You must have an existing cluster. For more information, see Create a cluster.

    Limitations

    • You cannot scale out a master node group. You can only scale out core and task node groups. By default, the new nodes inherit the configurations of existing nodes in the same group.

    • For Hadoop clusters, you cannot use this operation to scale out a node group that uses the Auto Scaling feature. For more information, see Configure Auto Scaling (Hadoop clusters only).

    Precautions

    If you have manually upgraded the StarRocks service in your cluster, subsequent scale-out operations can cause version inconsistencies. To maintain system stability, we recommend migrating to EMR Serverless StarRocks.

    EMR Serverless StarRocks is seamlessly compatible with open-source StarRocks and provides automatic software version upgrades, eliminating the complexity and risks of manual version management.

    Procedure

    Important

    Scale-out operations do not restart application processes on existing nodes.

    1. Go to the node management page.

      1. Log on to the EMR console.

      2. In the top navigation bar, select a region and a resource group as needed.

      3. In the Actions column of the target cluster, click Nodes.

    2. On the Nodes page, click Scale Out in the Actions column of the target node group.

    3. In the Scale Out dialog box, configure the parameters based on your business requirements.

      Parameter

      Description

      Node Group Name

      The node group name.

      Node Type

      The node group type.

      Current Instance Type

      The instance information of the node group.

      Billing Method

      During scale-out, the billing method is inherited from the existing node group.

      Pay-as-you-go clusters support new node groups that contain pay-as-you-go or preemptible instances.

      Subscription clusters support new node groups that contain subscription, pay-as-you-go, or preemptible instances.

      For the subscription billing method, you can configure whether to enable Auto-renewal and set the Duration for the new nodes.

      Note

      After you enable Auto-renewal, new subscription nodes added by scaling out are automatically renewed 7 days before they expire. The default renewal period is one month. You can later adjust the renewal duration and enable or disable the feature in the automatic renewal settings.

      vSwitch

      The vSwitch of the node group.

      Current Quantity

      The number of instances in the node group.

      Added Instances

      Use the arrows or enter a number to specify how many instances to add to the node group.

      Cluster Expiration Time

      The expiration time of the subscription cluster.

      Terms of Service

      Read and agree to the terms of service, then select the checkbox.

    4. After you configure the parameters, click OK.

      To log on to the newly scaled-out node, see Log on to a cluster.

    Next steps

    If you scale out a core node group that uses local HDFS for storage and the HDFS data is imbalanced after the scale-out is complete, use HDFS Balancer to rebalance the data in the DataNodes. For more information, see HDFS Balancer.

    Related documents