About Scaling an Oracle SOA Cloud Service Node Up or Down

You can scale only the Administration Server node and Managed Server nodes in a WebLogic Server cluster. Oracle SOA Cloud Service does not support scaling for other nodes in a service instance, such as the load balancer node.

You must scale each node in a cluster individually. You cannot scale all nodes in a cluster in a single operation.

Note:

Changing the compute shape to a higher value increases the billing of the Oracle SOA Cloud Service instance.

Oracle SOA Cloud Service provides a set of compute shapes that are optimized for different use cases. Choose from a set of all-purpose and memory-intensive shapes. The larger the compute shape, the greater the processing power:

• To meet the demands of heavier workloads, scale up the compute shape of a node by choosing a larger compute shape.

  For example, changing the compute shape from OC1M to OC2M or from VMStandard2.1 to VMStandard2.2 doubles the capacity of the node from one OCPU to two OCPUs and doubles the amount of RAM allocated to the node.

• To save costs if the workload is lightened, scale down the compute shape of a node by choosing a smaller compute shape.

  For example, changing the compute shape from OC2M to OC1M or from VMStandard2.2 to VMStandard2.1 reduces the capacity of the node by half from two OCPUs to one OCPU and reduces the amount of RAM allocated to the node by half.

Note:

To optimize performance and balance the load on Managed Sever instances correctly, ensure that the compute shapes of all nodes in a cluster are the same. When routing requests to Managed Server instances, the load balancer treats all Managed Server instances as being equivalent.

For steps to scale an Oracle SOA Cloud Service node, see Scale an Oracle SOA Cloud Service Node Up or Down.