Tair (Redis® OSS-Compatible):Enable automatic scale-up

Billing

You are charged only for the upgraded instance type. For more information, see configuration change.

How it works

After you enable automatic scale-up, when the average memory usage of an instance reaches the specified threshold within the observation window, DAS automatically scales up the instance. This operation upgrades the instance to the next larger instance type, for example, from 1 GB to 2 GB. After the scale-up is complete, DAS continues to monitor the memory usage. If the conditions for automatic scale-up are met again, DAS scales up the instance until it reaches the maximum instance type that you specified.

However, DAS does not automatically scale down an instance. If you enable the subscription service, when the average memory usage of the instance in the observation window drops below 30%, DAS sends you a scale-down recommendation through your configured channel, such as email. You can then manually scale down the instance to improve resource utilization. For more information about the subscription service, see the procedure in this topic.

Potential impacts

• The impact on connections during an automatic scale-up varies by instance type:

  • Cloud Disk Edition instance: If sufficient host resources are available, no connection flicker occurs during the scale-up. If a migration is required due to insufficient host resources, you may experience one or two connection flickers, each lasting less than 30 seconds.

  • Classic Edition instance: The scale-up process requires a migration, which causes one or two connection flickers, each lasting less than 30 seconds.

• To ensure the new instance can quickly synchronize incremental data from the original instance, the instance enters a read-only state for up to 1 minute during the configuration change.

• To ensure better performance and stability, if the instance uses an old minor version, the system automatically upgrades it to the latest minor version during the configuration change.

Procedure

1. Log on to the console and go to the Instances page. In the top navigation bar, select the region in which the instance is deployed. Then, find the instance and click its ID.

2. In the Configuration Information section, turn on the Auto Scaling switch.

3. In the DAS console dialog box that appears, configure automatic scale-up and event subscription settings.

   1. On the Autonomous Function Settings > Optimization And Flow Control tab, select the Auto Scaling checkbox and configure its parameters.

      Parameter	Description
      Average memory usage is not less than	The average memory usage threshold that triggers an automatic scale-up. Available in 10% increments from 50% to 90%.
      Maximum instance type	The maximum instance type available for scale-up. If the threshold is met, DAS upgrades the instance type one level at a time (for example, from 1 GB to 2 GB) and continues to monitor memory usage. If the threshold is met again within the observation window, the scale-up process repeats until the instance reaches this maximum instance type.
      Observation window	The duration of the observation window, in minutes. The maximum is 30 minutes. Note For example, if you set the threshold to 70%, the maximum instance type to 64 GB, and the observation window to 30 minutes, the system scales up the instance whenever its average memory usage is 70% or more within any 30-minute window, until the instance reaches the 64 GB instance type.

   2. Click OK.

      You are returned to the Tair console. The Automatic Scale-up status for the instance is now Enabled.

      If no alert template is configured for the instance, you can configure alert subscriptions to receive notifications for automatic scale-up events.

   3. Select an alert template and click Template Selection Completed, Next.

      The system recommends an alert template and adds the necessary alert rules. Configure the settings as prompted.

      Note

      • If you have already configured an alert template for the instance, follow the system prompts to add alert rules for the relevant autonomous events to the alert template.

      • If you want to configure your own alert templates and alert rules, see Configure alert templates and Configure alert rules.

   4. Select an alert contact group and click Contact Group Selection Completed, Next.

      To add contact groups or contacts, see Manage alert contacts.

   5. Click Submit Configuration and confirm the alert configuration in the dialog box that appears.

      This completes the alert subscription setup.

FAQ

• Q: After I enable automatic scale-up, the memory usage is very high (for example, 82%). Why doesn't the instance scale up?

  A: Check the threshold you set for average memory usage. For example, if the threshold is set to 90%, an automatic scale-up occurs only when the average memory usage of the instance reaches 90% or higher during the observation window.

• Q: Why is the automatic scale-up feature disabled after I enable read/write splitting?

  A: Currently, only the standard architecture supports automatic scale-up. Therefore, enabling read/write splitting disables the automatic scale-up feature.

Related documents

Redis Open-Source Edition instances that use the cloud-native cluster architecture also support automatic shard addition. For more information, see Automatic shard addition.