This page lists the specifications for read/write splitting instances of ApsaraDB for Redis Open-Source Edition, including memory, CPU cores, read-only nodes, bandwidth, maximum connections, and queries per second (QPS) reference values.
Read/write splitting instances are discontinued and no longer available for purchase. To evaluate current Redis instance types, see cluster instances.
Instance specifications
In the following table, "1 node" means the instance has a single data shard. For example, the 1 GB Read/write Splitting (1 Shard, 3 Read-only Nodes) instance type has one data shard with three read-only nodes.
| Specification | InstanceClass (for API calls) | CPU cores | Read-only nodes | Bandwidth (MB/s) | Maximum connections | QPS reference value |
|---|---|---|---|---|---|---|
| 1 GB read/write splitting (1 shard, 1 read-only node) | redis.logic.splitrw.small.1db.1rodb.4proxy.default | 4 | 1 | 96 | 20,000 | 200,000 |
| 1 GB read/write splitting (1 shard, 3 read-only nodes) | redis.logic.splitrw.small.1db.3rodb.4proxy.default | 8 | 3 | 192 | 40,000 | 400,000 |
| 1 GB read/write splitting (1 shard, 5 read-only nodes) | redis.logic.splitrw.small.1db.5rodb.6proxy.default | 12 | 5 | 288 | 60,000 | 600,000 |
| 2 GB read/write splitting (1 shard, 1 read-only node) | redis.logic.splitrw.mid.1db.1rodb.4proxy.default | 4 | 1 | 192 | 20,000 | 200,000 |
| 2 GB read/write splitting (1 shard, 3 read-only nodes) | redis.logic.splitrw.mid.1db.3rodb.4proxy.default | 8 | 3 | 384 | 40,000 | 400,000 |
| 2 GB read/write splitting (1 shard, 5 read-only nodes) | redis.logic.splitrw.mid.1db.5rodb.6proxy.default | 12 | 5 | 576 | 60,000 | 600,000 |
| 4 GB read/write splitting (1 shard, 1 read-only node) | redis.logic.splitrw.stand.1db.1rodb.4proxy.default | 4 | 1 | 192 | 20,000 | 200,000 |
| 4 GB read/write splitting (1 shard, 3 read-only nodes) | redis.logic.splitrw.stand.1db.3rodb.4proxy.default | 8 | 3 | 384 | 40,000 | 400,000 |
| 4 GB read/write splitting (1 shard, 5 read-only nodes) | redis.logic.splitrw.stand.1db.5rodb.6proxy.default | 12 | 5 | 576 | 60,000 | 600,000 |
| 8 GB read/write splitting (1 shard, 1 read-only node) | redis.logic.splitrw.large.1db.1rodb.4proxy.default | 4 | 1 | 192 | 20,000 | 200,000 |
| 8 GB read/write splitting (1 shard, 3 read-only nodes) | redis.logic.splitrw.large.1db.3rodb.4proxy.default | 8 | 3 | 384 | 40,000 | 400,000 |
| 8 GB read/write splitting (1 shard, 5 read-only nodes) | redis.logic.splitrw.large.1db.5rodb.6proxy.default | 12 | 5 | 576 | 60,000 | 600,000 |
| 16 GB read/write splitting (1 shard, 1 read-only node) | redis.logic.splitrw.2xlarge.1db.1rodb.4proxy.default | 4 | 1 | 192 | 20,000 | 200,000 |
| 16 GB read/write splitting (1 shard, 3 read-only nodes) | redis.logic.splitrw.2xlarge.1db.3rodb.4proxy.default | 8 | 3 | 384 | 40,000 | 400,000 |
| 16 GB read/write splitting (1 shard, 5 read-only nodes) | redis.logic.splitrw.2xlarge.1db.5rodb.6proxy.default | 12 | 5 | 576 | 60,000 | 600,000 |
| 32 GB read/write splitting (1 shard, 1 read-only node) | redis.logic.splitrw.4xlarge.1db.1rodb.4proxy.default | 4 | 1 | 192 | 20,000 | 200,000 |
| 32 GB read/write splitting (1 shard, 3 read-only nodes) | redis.logic.splitrw.4xlarge.1db.3rodb.4proxy.default | 8 | 3 | 384 | 40,000 | 400,000 |
| 32 GB read/write splitting (1 shard, 5 read-only nodes) | redis.logic.splitrw.4xlarge.1db.5rodb.6proxy.default | 12 | 5 | 576 | 60,000 | 600,000 |
| 64 GB read/write splitting (1 shard, 1 read-only node) | redis.logic.splitrw.8xlarge.1db.1rodb.4proxy.default | 4 | 1 | 192 | 20,000 | 200,000 |
| 64 GB read/write splitting (1 shard, 3 read-only nodes) | redis.logic.splitrw.8xlarge.1db.3rodb.4proxy.default | 8 | 3 | 384 | 40,000 | 400,000 |
| 64 GB read/write splitting (1 shard, 5 read-only nodes) | redis.logic.splitrw.8xlarge.1db.5rodb.6proxy.default | 12 | 5 | 576 | 60,000 | 600,000 |
Notes about CPU cores
To maintain service stability, the system reserves one CPU core per shard or read-only node for background tasks. This applies to both cluster instances and read/write splitting instances.
Calculation rules for bandwidth values
Aggregate bandwidth: Each value in the table represents the maximum bandwidth for the entire instance — the combined bandwidth across all shards and nodes, not a per-node limit.
Maximum cap: The total bandwidth for a read/write splitting instance cannot exceed 2,048 Mbit/s. Adding more shards or nodes does not increase bandwidth beyond this limit.
Bidirectional: The bandwidth applies equally to upstream and downstream traffic. For example, if the bandwidth of an instance is 96 Mbit/s, the upstream and downstream bandwidths of the instance are both 96 Mbit/s.
Scope: The bandwidth limits of Tair and Redis refer to the data transfer capacity of individual shards within their respective distributed systems. The limits are generally independent of the network connection types used by clients to connect to the shards.
To handle traffic spikes, you can increase the instance bandwidth on demand. See Manually increase the bandwidth of an instance.
For common bandwidth questions, see FAQ about bandwidth.