Updated on 2024-07-11 GMT+08:00

Elastic Scaling of Queues

Prerequisites

Elastic scaling can be performed for a newly created pay-per-use queue only when there were jobs running in this queue.

Notes and Constraints

  • Queues with 16 CUs do not support scale-out or scale-in.
  • Queues with 64 CUs do not support scale-in.
  • Only queues whose billing mode is Pay-per-use/By CUH and Pay-per-use/Dedicated resource mode support elastic scaling.
  • If Status of queue xxx is assigning, which is not available is displayed on the Elastic Scaling page, the queue can be scaled only after the queue resources are allocated.
  • If there are not enough physical resources, a queue may not be able to scale out to the desired target size.
  • The system does not guarantee that a queue will be scaled in to the desired target size. Typically, the system checks the resource usage before scaling in the queue to determine if there is enough space for scaling in. If the existing resources cannot be scaled in according to the minimum scaling step, the queue may not be scaled in successfully or only partially.

    The scaling step may vary depending on the resource specifications, usually 16 CUs, 32 CUs, 48 CUs, 64 CUs, etc.

    For example, if the queue size is 48 CUs and job execution uses 18 CUs, the remaining 30 CUs do not meet the requirement for scaling in by the minimum step of 32 CUs. If a scaling in task is executed, it will fail.

Scaling Out

If the current queue specifications do not meet service requirements, you can add the number of CUs to scale out the queue.

Scale-out is time-consuming. After you perform scale-out on the Elastic Scaling page of DLI, wait for about 10 minutes. The duration is related to the CU amount to add. After a period of time, refresh the Queue Management page and check whether values of Specifications and Actual CUs are the same to determine whether the scale-out is successful. Alternatively, on the Job Management page, check the status of the SCALE_QUEUE SQL job. If the job status is Scaling, the queue is being scaled out.

The procedure is as follows:

  1. On the left of the DLI management console, click Resources > Queue Management.
  2. Select the queue to be scaled out and choose More > Elastic Scaling in the Operation column.
  3. On the displayed page, select Scale-out for Operation and set the scale-out amount.
    Figure 1 Elastic scale-out
  4. Click OK.

Scaling In

If the current queue specifications are too much for your computing service, you can reduce the number of CUs to scale in the queue.

  • Scale-in is time-consuming. After you perform scale-in on the Elastic Scaling page of DLI, wait for about 10 minutes. The duration is related to the CU amount to reduce. After a period of time, refresh the Queue Management page and check whether values of Specifications and Actual CUs are the same to determine whether the scale-in is successful. Alternatively, on the Job Management page, check the status of the SCALE_QUEUE SQL job. If the job status is Scaling, the queue is being scaled in.
  • By default, the minimum number of CUs is 16. That is, when the queue specifications are 16 CUs, you cannot scale in the queue.

The procedure is as follows:

  1. On the left of the DLI management console, click Resources > Queue Management.
  2. Select the queue to be scaled out, click More in the Operation column, and select Elastic Scaling.
  3. On the displayed page, select Scale-in for Operation and set the scale-in amount.
    Figure 2 Manual scale-in
  4. Click OK.