Change memory capacity and instance architecture - Tair (Redis® OSS-Compatible)

Tair (Redis OSS-compatible) lets you flexibly change instance configurations, including architecture, memory capacity, number of shards, and number of replica nodes, to meet evolving business demands for performance and capacity.

What to expect

For all configuration changes:

You do not need to modify your application code to adapt to the changes. The endpoints, database accounts, passwords, and whitelist settings of the instance remain unchanged.
Instance data is preserved. In the rare event that the original primary node fails during switchover, there is a theoretical risk of losing a small amount of unsynchronized data.

The following table lists the supported configuration changes and the subsequent impacts.

Change	Impact
Change the instance architecture from standard (master-replica) to cluster	One to two transient disconnections, each lasting less than 30 seconds. The instance is read-only for about one minute.
Change the instance architecture from cluster to standard (master-replica)	One to two transient disconnections, each lasting less than 30 seconds. The instance becomes read-only for approximately one minute.
Change the number of cluster shards	Possible one to two transient disconnections, each lasting less than 30 seconds. The instance becomes read-only for approximately one minute.
Enable or disable read/write splitting	One transient disconnection that lasts for a few seconds.
Add or remove replica nodes	None.
Change the deployment mode from classic to cloud-native	One transient disconnection lasting less than 30 seconds. The instance becomes read-only for approximately one minute.

FAQ

How can I check if an instance has been scaled out?

Filter for tasks where Status is Successful on the Task Center page and check for tasks where Task Type is Modify Instance Specification or Instance. You can also query this information by calling DescribeHistoryTasks.

What may cause a configuration change to fail?

A change may fail for two main reasons:

The instance contains large keys. Identify and delete all large keys before changing the instance configuration. See Use the offline key analysis feature for details.

The used memory of the instance exceeds 80% of the memory capacity after a downgrade. For example, if you have a DRAM-based instance with 2 GB of memory in use, you cannot downgrade its memory to less than 2.5 GB.

How do I change the storage type of a Tair (Enterprise Edition) instance?

You cannot change it because the storage types are not compatible.

Can I upgrade only the CPU specifications of an instance?

No, you cannot upgrade only the CPU specifications of Tair (including Redis Open-Source Edition) instances. To increase the CPU cores of an instance:

Change the instance architecture from standard to architecture or enable read/write splitting.
Increase the number of read-only nodes for a instance with read/write splitting enabled.
Increase the number of shards for a cluster instance.

For more information, see How to upgrade the CPU specifications of an instance and Instance types and FAQ.

How do I change a High-Availability (dual-replica) instance to a Standalone instance?

Tair does not support changing a High Availability (HA) instance to a Standalone instance because Standalone instances do not provide data reliability guarantees.

If needed, purchase a Standalone instance and use Data Transmission Service (DTS) to migrate data from your HA instance to it. For more information, see Migrate data between Tair (Redis OSS-compatible) instances.

Do I need to pause read and write operations during a configuration change?

No. However, perform changes during off-peak hours to minimize business impact. The instance may become read-only for about one minute and experience one or two transient disconnections, each lasting less than 30 seconds. For more information about the specific impacts of each change, see What to expect.

When I change a standard instance to a cluster instance or change the number of shards for a cluster instance, is the data automatically migrated to each shard?

Yes, the system automatically migrates and rebalances the data across all shards.

How long does a configuration change take?

The time varies based on factors such as network conditions, request volume, and data size, making it hard to predicted.

You can monitor the task progress by clicking the icon in the upper-right corner of the instance details page.

Will a configuration change cause any backup sets to be lost?

Configuration changes do not delete backup sets. However, when you reduce the number of shards for a classic cluster instance or change it to a standard instance, the mapping between historical backup sets and instance nodes changes.

To find historical backup sets in this scenario, search for the relevant backup sets by their creation time or backup set ID.

You can download the historical backup set (RDB file), parse it, and import the data into the new instance to restore the instance.

Why do the instance configurations remain unchanged on the console after I change them?

There may be a brief delay before the new configurations appear on the console due to caching. Please wait a few minutes and then refresh the page.

Related API operations

API operation	Description
ModifyInstanceSpec	Changes the specifications of an instance.