Tair (Redis® OSS-Compatible):FAQ about bandwidth

Is the bandwidth of an instance limited? What is the maximum bandwidth?

Tair (and Redis Open-Source Edition) instances have bandwidth limits. The maximum bandwidth varies based on the instance type.

You can query the maximum bandwidth of an instance using one of the following methods:

• Go to the Instance Information page in the Tair console or call the DescribeIntranetAttribute operation to query the bandwidth limit.

• Find the bandwidth limit for your instance type in Instance types and FAQ.

Does bandwidth refer to internal bandwidth or public bandwidth?

The bandwidth limit of a Tair instance is the bandwidth of a shard and is independent of the network connection type.

For example, if the bandwidth limit for a tair.rdb.1g instance is 96 MB/s, the total available bandwidth for a single shard is 96 MB/s. This applies regardless of whether connections are made over an internal network or the internet.

Does bandwidth refer to upstream bandwidth or downstream bandwidth?

The bandwidth value applies to the upstream and downstream bandwidths. For example, if the bandwidth of an instance is 192 Mbps (24 MB/s), the upstream and downstream bandwidths of the instance are both 192 Mbps (24 MB/s).

How is the bandwidth of a cluster or read/write splitting instance calculated?

The total bandwidth of a cluster architecture or read/write splitting instance is the sum of the bandwidths of all its shards or replicas.

For example, if an instance has four shards of the redis.shard.with.proxy.small.ce type, and the bandwidth of each shard is 24 MB/s, the total bandwidth of the instance is 96 MB/s.

How do I upgrade bandwidth?

• For cluster architecture or read/write splitting instances, you can increase the total bandwidth by adding more shards or read replicas. For more information, see Adjust the number of shards for a cluster and Enable read/write splitting.

  Note

  The maximum bandwidth of a cluster instance in proxy mode or a read/write splitting instance is 2,048 Mbit/s. After the upper limit is reached, the bandwidth cannot be increased even if you add more shards to the instance. Cluster instances in direct connection mode have no limits on the overall bandwidth.

• You can adjust the bandwidth of all instance types except for Tair ESSD-based instances. For more information, see Manually increase the bandwidth of an instance.

• You can enable automatic bandwidth scaling for Redis Open-Source Edition instances and Tair Memory-optimized instances that are deployed in classic mode. For more information, see Enable bandwidth Auto Scaling.

Am I charged bandwidth and traffic fees for using a Tair instance?

Data transfer fees: Tair does not charge data transfer fees.

Bandwidth fees: You are not charged for bandwidth if you do not manually adjust the instance bandwidth or enable bandwidth Auto Scaling. Bandwidth fees apply if you adjust the bandwidth. For more information about billing, see Billable items.

Why is the bandwidth of my instance different from the documented bandwidth?

Cause:

If your instance is a Redis Open-Source Edition cloud-native instance that was created before June 20, 2023, its bandwidth may not match the documented bandwidth. This is because Alibaba Cloud adjusted the default bandwidth of Redis Open-Source Edition cloud-native instances on June 20, 2023. The documentation describes the new bandwidth after this adjustment.

Solution:

The adjustment applies only to Redis Open-Source Edition cloud-native instances that are created or modified after June 20, 2023.

To apply the increased default bandwidth to your existing instances, change the configurations of the instances on or after June 20, 2023. For example, you can upgrade the specifications or change the architecture for the instances.

For more information, see [Notice] Default bandwidth is increased for Redis Community Edition cloud-native instances.

Which key's operation is using the most bandwidth?

Tair does not provide a feature to identify which key or command uses the most bandwidth.

To resolve high traffic usage, see Troubleshoot high traffic usage on an instance.