This topic describes the specifications of read/write splitting instances of ApsaraDB for Redis Community Edition. These specifications include the memory capacity, maximum number of concurrent connections to each instance, maximum bandwidth, and queries per second (QPS) reference value.

## Instance types

**Note**The following table describes the instance types that contain only one shard. For example, a 1 GB read/write splitting instance (1 shard, 3 read replicas) indicates that the instance has one shard that contains three read replicas.

Specification | InstanceClass (used in API operations) | Number of read replicas | Bandwidth (Mbit/s) | Maximum number of new connections per second | Maximum number of connections | QPS reference value |
---|---|---|---|---|---|---|

1 GB read/write splitting instance (1 shard, 1 read replica) | redis.logic.splitrw.small.1db.1rodb.4proxy.default | 1 | 96 | 20000 | 20000 | 200000 |

1 GB read/write splitting instance (1 shard, 3 read replicas) | redis.logic.splitrw.small.1db.3rodb.4proxy.default | 3 | 192 | 40000 | 40000 | 400000 |

1 GB read/write splitting instance (1 shard, 5 read replicas) | redis.logic.splitrw.small.1db.5rodb.6proxy.default | 5 | 288 | 50000 | 60000 | 600000 |

2 GB read/write splitting instance (1 shard, 1 read replica) | redis.logic.splitrw.mid.1db.1rodb.4proxy.default | 1 | 192 | 20000 | 20000 | 200000 |

2 GB read/write splitting instance (1 shard, 3 read replicas) | redis.logic.splitrw.mid.1db.3rodb.4proxy.default | 3 | 384 | 40000 | 40000 | 400000 |

2 GB read/write splitting instance (1 shard, 5 read replicas) | redis.logic.splitrw.mid.1db.5rodb.6proxy.default | 5 | 576 | 50000 | 60000 | 600000 |

4 GB read/write splitting instance (1 shard, 1 read replica) | redis.logic.splitrw.stand.1db.1rodb.4proxy.default | 1 | 192 | 20000 | 20000 | 200000 |

4 GB read/write splitting instance (1 shard, 3 read replicas) | redis.logic.splitrw.stand.1db.3rodb.4proxy.default | 3 | 384 | 40000 | 40000 | 400000 |

4 GB read/write splitting instance (1 shard, 5 read replicas) | redis.logic.splitrw.stand.1db.5rodb.6proxy.default | 5 | 576 | 50000 | 60000 | 600000 |

8 GB read/write splitting instance (1 shard, 1 read replica) | redis.logic.splitrw.large.1db.1rodb.4proxy.default | 1 | 192 | 20000 | 20000 | 200000 |

8 GB read/write splitting instance (1 shard, 3 read replicas) | redis.logic.splitrw.large.1db.3rodb.4proxy.default | 3 | 384 | 40000 | 40000 | 400000 |

8 GB read/write splitting instance (1 shard, 5 read replicas) | redis.logic.splitrw.large.1db.5rodb.6proxy.default | 5 | 576 | 50000 | 60000 | 600000 |

16 GB read/write splitting instance (1 shard, 1 read replica) | redis.logic.splitrw.2xlarge.1db.1rodb.4proxy.default | 1 | 192 | 20000 | 20000 | 200000 |

16 GB read/write splitting instance (1 shard, 3 read replicas) | redis.logic.splitrw.2xlarge.1db.3rodb.4proxy.default | 3 | 384 | 40000 | 40000 | 400000 |

16 GB read/write splitting instance (1 shard, 5 read replicas) | redis.logic.splitrw.2xlarge.1db.5rodb.6proxy.default | 5 | 576 | 50000 | 60000 | 600000 |

32 GB read/write splitting instance (1 shard, 1 read replica) | redis.logic.splitrw.4xlarge.1db.1rodb.4proxy.default | 1 | 192 | 20000 | 20000 | 200000 |

32 GB read/write splitting instance (1 shard, 3 read replicas) | redis.logic.splitrw.4xlarge.1db.3rodb.4proxy.default | 3 | 384 | 40000 | 40000 | 400000 |

32 GB read/write splitting instance (1 shard, 5 read replicas) | redis.logic.splitrw.4xlarge.1db.5rodb.6proxy.default | 5 | 576 | 50000 | 60000 | 600000 |

64 GB read/write splitting instance (1 shard, 1 read replica) | redis.logic.splitrw.8xlarge.1db.1rodb.4proxy.default | 1 | 192 | 20000 | 20000 | 200000 |

64 GB read/write splitting instance (1 shard, 3 read replicas) | redis.logic.splitrw.8xlarge.1db.3rodb.4proxy.default | 3 | 384 | 40000 | 40000 | 400000 |

64 GB read/write splitting instance (1 shard, 5 read replicas) | redis.logic.splitrw.8xlarge.1db.5rodb.6proxy.default | 5 | 576 | 50000 | 60000 | 600000 |

## Calculation rules for bandwidth

- If network resources are sufficient, the bandwidth is unlimited. However, if network resources are insufficient, the maximum instance bandwidth is limited.
- The bandwidth value in the table is the bandwidth of the instance. It is the sum of the bandwidth of all shards or nodes on the instance.
- The upper limit of the total bandwidth for a read/write splitting instance (as shown in the table) is 2,048 Mbit/s. After the upper limit is reached, the bandwidth cannot be increased even if the number of shards or nodes is increased.
- The bandwidth value applies to the upstream and downstream bandwidths. If the bandwidth of an instance is 10 Mbit/s, the upstream and downstream bandwidths of the instance are both 10 Mbit/s.
- The bandwidth listed in the table is the internal bandwidth of the ApsaraDB for Redis instance. The Internet bandwidth is determined by the internal bandwidth and is limited by the bandwidth of the connection between the instance and a client. We recommend that you connect to the instance over an internal network to maximize the performance.

## Calculation rules for connections

- The following example shows the maximum number of new connections that can be created per second: The maximum number of new connections per second of an instance is 10,000. The maximum number of connections is 50,000. The actual number of connections at the Nth second after the instance starts to run is 12,000. Then, at the (N+1)th second, the number of connections is 22,000 (12,000 + 10,000).
- The maximum number of new connections per second and the maximum number of connections
in the preceding table both refer to the total number of connections of the instance
in proxy mode. The maximum number of connections is 500,000 or less. After the value
is reached, the maximum number of connections is not increased even if the number
of shards or nodes is increased.
**Note**If you enable direct connection for a cluster instance, the connectivity performance of the instance degrades because the proxy node is bypassed. For cluster instances of Community Edition, the maximum number of connections for a single shard is 10,000, and the maximum number of new connections per second is 2,000. For cluster instances of Enhanced Edition, the maximum number of connections for an individual shard is 30,000, the maximum number of new connections per second is 2,000.

## Calculation rules for connections

The following example shows the maximum number of new connections that can be added per second: The maximum number of new connections per second of an instance is 10,000. The maximum number of connections is 50,000. The actual number of connections at the Nth second after the instance starts to run is 12,000. Then, at the (N+1)th second, the number of connections is 22,000 (12,000 + 10,000).