Tair (Redis OSS-compatible) does not allow you to adjust the number of CPU cores. You can change the instance architecture or add more nodes or shards to improve the overall performance of an instance. If you have already identified and optimized factors such as CPU-intensive commands, hotkeys, and large keys, and confirmed that no abnormal traffic affecting the performance exists, but the CPU load remains high, you can refer to this topic to improve the overall CPU performance of the instance.
Confirm the edition, deployment mode, and specifications of an instance
The CPU specifications supported by an instance vary based on the edition, deployment mode, and specifications of the instance. Before you upgrade the CPU specifications, check the information about the instance.
Log on to the console and go to the Instances page. In the top navigation bar, select the region in which the instance that you want to manage resides. Then, find the instance and click the instance ID.
In the Basic Information and Configuration Information sections of the Instance Information page, view the following information: Edition, Specifications, and Deployment Mode.
NoteFor different instances, the UI terms and positions of Edition, Specifications, and Deployment Mode may slightly vary. The actual UI terms in the console prevail.
For a cloud-native instance, click Node Management in the left-side navigation pane to check whether read/write splitting is enabled.
Standard instances
Redis Open-Source Edition instances and Tair DRAM-based and persistent memory-optimized instances
Confirm the series type and architecture of an instance
On the Instance Information page, make sure that the instance meets the following requirements:
The instance uses the standard architecture.
If the Edition parameter is set to Tair (Enterprise Edition), the series type is DRAM or Persistent Memory.
Methods to improve CPU performance
The CPU specifications of a single shard in a standard instance cannot be changed. You cannot add more cores to scale up the CPU capacity of a shard. However, you can change the architecture to improve the overall performance of the instance:
If your workloads are read/write balanced or write-heavy, you can switch to the cluster architecture. By increasing the number of shards, you can improve the overall CPU performance of the instance. For more information, see Change the configurations of an instance.
If your workloads are read-heavy, you can switch to the read/write splitting architecture. You can enable read/write splitting and increase the number of read replicas to improve the overall CPU performance of the instance.
For information about how to enable read/write splitting for a cloud-native instance, see Enable read/write splitting.
For information about how to enable read/write splitting for a classic instance, see Change the configurations of an instance.
NoteDue to the inherent latency in synchronization between the master node and read replicas, the read/write splitting architecture is suitable for applications that can tolerate a certain level of dirty reads. If your business cannot tolerate dirty reads, we recommend that you switch to the cluster architecture.
Tair ESSD/SSD-based instances
Confirm the series type and architecture of an instance
On the Instance Information page of the instance, make sure that the Edition parameter is set to Tair (Enterprise Edition) and the series type is ESSD.
Methods to improve CPU performance
You can change the shard specifications to improve CPU performance. For more information, see Change the configurations of an instance.
Cluster instances
Confirm the series type and architecture of an instance
On the Instance Information page, make sure that the instance uses the cluster architecture.
Methods to improve CPU performance
The CPU specifications of a single shard in a cluster instance cannot be changed. You cannot add more cores to scale up the CPU capacity of a shard. However, you can increase the number of shards to improve the overall performance of the instance.
For information about how to increase the number of shards for a cloud-native instance, see Adjust the number of shards for a cluster instance.
For information about how to increase the number of shards for a classic instance, see Change the configurations of an instance.
If a bandwidth plan has been upgraded for a cluster instance, monitor the traffic of the instance after you increase the number of shards, and determine whether to upgrade the bandwidth plan for the newly added shards.
Read/write splitting instances
Confirm the series type and architecture of an instance
On the Instance Information page, make sure that the instance uses the read/write splitting architecture. Alternatively, you have turned on the Read/Write Splitting switch on the Node Management page.
Methods to improve CPU performance
The CPU specifications of a read/write splitting instance cannot be changed. You cannot add more cores to scale up the CPU capacity. However, you can increase the number of read replicas to improve the overall performance of the instance.
For information about how to increase the number of read replicas for a cloud-native instance, see Enable read/write splitting.
For information about how to increase the number of read replicas for a classic instance, see Change the configurations of an instance.
References
For information about how to troubleshoot and address high CPU utilization, see Troubleshoot high CPU utilization on an instance.
For information about the performance of Redis Open-Source Edition and Tair (Enterprise Edition), see Performance White Paper.
For information about the instance type, CPU cores, memory, bandwidth, and maximum allowed number of connections of Redis Open-Source Edition and Tair (Enterprise Edition) instances, see Instance specifications.