To cope with burst or expected traffic peaks, you can temporarily increase the network bandwidth on the Instance Information page of the console.

Background information

If big keys are used for business or the traffic generated during peak hours exceeds the bandwidth, Redis nodes may be blocked, which deteriorates the service performance. In this case, you can temporarily adjust the bandwidth. This buys more time for you to optimize the memory architecture and process big keys. You can also easily cope with the peak traffic generated during time-limited activities and focus on improving your business.

Bandwidth throttling adjustment rules:

  • The temporarily adjusted bandwidth is valid for at least seven days and expires at 00:00 after seven days. For example, if you adjust the bandwidth at 10:00 on January 1, the adjusted upper limit expires at 00:00 on January 9.
    Note For more information about how to permanently adjust bandwidth throttling, see Change the instance configuration.
  • After the adjustment, the bandwidth is doubled. For example, if the bandwidth for an instance is 10 MB/s, it is raised to 20 MB/s after the adjustment.
Notice The following operations may restore the bandwidth to the original value:
  • Upgrade the major version or minor version for instances of the cluster edition.
  • Upgrade or downgrade the configuration for instances of all editions or migrate instances of all editions across zones.


  1. Log on to the ApsaraDB for Redis console.
  2. In the upper-left corner of the top navigation bar, select the region where the target instance is located.
  3. On the Instance List page, click the target instance ID or Manage in the Action column for the target instance.
  4. On the Instance Information page, click Temporarily Adjust Bandwidth to the right of the Maximum Internal Bandwidth field in the Basic Information section.
    Note Before the adjustment, 96 MB/s in the Maximum Internal Bandwidth field for an instance of the cluster edition indicates the bandwidth for a single node in the cluster.