Database Autonomy Service (DAS) auto scaling monitors your database instances and adjusts their resources automatically based on policies you define. When a workload reaches a specified threshold, DAS either scales the instance or provides a scaling suggestion. Auto scaling has been gradually rolling out to all DAS users since February 18, 2022.
Scaling modes
DAS supports four scaling modes. Each targets a different resource type and trigger mechanism.
| Mode | What it scales | Trigger |
|---|---|---|
| Prediction-based auto scaling | Instance specifications | Predicted metric value (based on 10-day history) >= preset threshold |
| Scheduled auto scaling | Instance specifications | A predefined schedule |
| Automatic scaling for local resources | CPU cores (local host resources) | Average CPU utilization over an observation window >= preset threshold |
| Automatic storage expansion | Storage capacity | Used storage capacity >= preset threshold |
Choose a scaling mode
Select a mode based on the load pattern your database experiences:
Unpredictable load spikes — Use prediction-based auto scaling. DAS analyzes the last 10 days of historical data to forecast metric values for the next 24 hours and suggests scaling before thresholds are breached.
Predictable periodic demand — Use scheduled auto scaling. Set a fixed schedule to scale up before peak periods and scale back to original specifications afterward, keeping costs minimal during off-peak hours.
CPU bottlenecks on the local host — Use automatic scaling for local resources. DAS reallocates available CPU cores on the host to the database instance when average utilization crosses your threshold.
Storage running low — Use automatic storage expansion. DAS expands storage capacity as soon as used storage reaches your threshold, preventing service disruption.
Limitations
Review these constraints before configuring policies:
Only policies with no applied instances can be deleted. Cancel the policy on all instances first.
Modifying a policy affects all instances it is currently applied to.
Modifying a scheduled auto scaling policy while a task is in progress affects scale-back behavior:
If you change the Operation parameter during an execution period, DAS does not scale the instances back to their original specifications at scale-back time.
If you change Valid From and Recurrence during an execution period and the Scale-back Time changes as a result, DAS does not scale the instances back to their original specifications.
Deleting a scheduled auto scaling policy while a task is in progress prevents DAS from scaling instances back to their original specifications.
Create a policy
Log on to the DAS console.
In the left-side navigation pane, choose Resources > Auto Scaling Settings.
In the Auto Scaling Policies section, click Add Policy.
In the Add Policy panel, select a mode from the Mode drop-down list and configure the parameters. For parameter details, see the topic for your selected mode:
Apply a policy
Log on to the DAS console.
In the left-side navigation pane, choose Resources > Auto Scaling Settings.
In the Auto Scaling Policies section, find the policy and click Apply in the Actions column.
In the Apply Policies dialog box, select the target database instances and click the
icon.Click Confirm.
Cancel a policy
Log on to the DAS console.
In the left-side navigation pane, choose Resources > Auto Scaling Settings.
In the Auto Scaling Policies section, click the number in the Applied Instances column for the policy you want to cancel.
In the Instance Details dialog box, find the target instance and click Cancel in the Actions column.
Click OK.
Modify a policy
Modifying a policy affects all instances it is applied to. If you modify a scheduled auto scaling policy while a task is running, see Limitations for the impact on scale-back behavior.
Log on to the DAS console.
In the left-side navigation pane, choose Resources > Auto Scaling Settings.
In the Auto Scaling Policies section, find the policy and click Modify in the Actions column.
In the Update Policy panel, update the settings.
Click OK.
Delete a policy
Only policies with no applied instances can be deleted. If the policy is still applied to instances, cancel it first. Deleting a scheduled auto scaling policy while a task is running prevents scale-back — see Limitations.
Log on to the DAS console.
In the left-side navigation pane, choose Resources > Auto Scaling Settings.
In the Auto Scaling Policies section, delete the policy using one of the following methods:
Single policy: Find the policy and click Delete in the Actions column.
Multiple policies: Select the policies and click Batch Delete.