Single-replica cluster instances

Instance types

Each data shard (hereinafter referred to as shard) of a single-replica cluster instance has only one node.

The instance types offer various options for the number of shards: 8 shards, 16 shards, 32 shards, 64 shards, and 128 shards.

• 8 shards

  For this series of instance types, the number of shards is 8 and the number of CPU cores is 16 (2 CPU cores per shard).

  Specification Name	Instance type code (for API calls)	Memory per shard GB	Proxy mode Maximum connections	Direct connection mode Maximum Connections	Total bandwidth	Bandwidth per shard	QPS reference
  16 GB single-replica cluster instance	redis.sharding.basic.small.default	2	500,000	80,000	6,144 Mbps (768 MB/s)	768 Mbps (96 MB/s)	800,000
  32 GB single-replica cluster instance	redis.sharding.basic.mid.default	4	500,000	80,000	6,144 Mbps (768 MB/s)	768 Mbps (96 MB/s)	800,000
  64 GB single-replica cluster instance	redis.sharding.basic.large.default	8	500,000	80,000	6,144 Mbps (768 MB/s)	768 Mbps (96 MB/s)	800,000

• 16 shards

  For this series of instance types, the number of shards is 16 and the number of CPU cores is 32 (2 CPU cores per shard).

  Instance type name	Instance type code (for API calls)	Memory per shard (GB)	Proxy mode Maximum connections	Direct connection mode Maximum connections	Total bandwidth	Bandwidth per shard	QPS reference
  128 GB single-replica cluster instance	redis.sharding.basic.2xlarge.default	8	500,000	160,000	12,288 Mbps (1,536 MB/s)	768 Mbps (96 MB/s)	1,600,000
  256 GB single-replica cluster instance	redis.sharding.basic.4xlarge.default	16	500,000	160,000	12,288 Mbps (1,536 MB/s)	768 Mbps (96 MB/s)	1,600,000

• 32 shards

  For this series of instance types, the number of shards is 32 and the number of CPU cores is 64 (2 CPU cores per shard).

  Instance type name	Instance type code (for API calls)	Memory per shard (GB)	Proxy mode Maximum connections	Direct connection mode Maximum Connections	Total bandwidth	Bandwidth per shard	QPS reference
  512 GB single-replica cluster instance	redis.sharding.basic.8xlarge.default	16	500,000	320,000	16,384 Mbps (2,048 MB/s)	768 Mbps (96 MB/s)	3,200,000

• 64 shards

  For this series of instance types, the number of shards is 64 and the number of CPU cores is 128 (2 CPU cores per shard).

  Instance type name	Instance type code (for API calls)	Memory per shard (GB)	Proxy mode Maximum connections	Direct connection mode Maximum connections	Total bandwidth	Bandwidth per shard	QPS reference
  1 TB single-replica cluster instance	redis.sharding.basic.16xlarge.default	16	500,000	640,000	16,384 Mbps (2,048 MB/s)	768 Mbps (96 MB/s)	6,400,000

• 128 shards

  For this series of instance types, the number of shards is 128 and the number of CPU cores is 256 (2 CPU cores per shard).

  Instance type name	Instance type code (for API calls)	Memory per shard (GB)	Proxy mode Maximum connections	Direct connection mode Maximum connections	Total bandwidth	Bandwidth per shard	QPS reference
  2 TB single-replica cluster instance	redis.sharding.basic.32xlarge.default	16	500,000	1,280,000	16,384 Mbps (2,048 MB/s)	768 Mbps (96 MB/s)	12,800,000

CPU cores

Each instance has 2 CPU cores. One core is reserved for background tasks to maintain service stability. For cluster or read/write splitting instances, one core is reserved per data shard or read-only node.

Bandwidth calculation rules

• The total bandwidth values in the tables are the bandwidth of the entire instance, which is the sum of the bandwidth of all shards or nodes in the instance. The bandwidth limit of each shard is as follows:

  • Redis Open-Source Edition: If the memory of a single shard is less than or equal to 1 GB, the maximum bandwidth of each shard is 384 Mbps (48 MB/s). If the memory of a single shard is greater than 1 GB, the maximum bandwidth of each shard is 768 Mbps (96 MB/s).

  • Tair (Enterprise Edition): The maximum bandwidth of each shard is 768 Mbps (96 MB/s).

• When a cluster instance is connected by using the default proxy endpoint, the maximum bandwidth is 16,384 Mbps (2,048 MB/s). The bandwidth does not increase beyond this limit, even if you increase the number of shards.

  To handle extremely high network traffic, you can enable the direct connection mode. For more information, see Enable direct connection (applicable only to the cluster architecture). After you enable the direct connection mode:

  • Maximum connections: the maximum number of connections per shard multiplied by the number of shards. The maximum number of connections per shard is 10,000 for Redis Open-Source Edition and 30,000 for Tair (Enterprise Edition).

  • Overall bandwidth limit: the maximum bandwidth per shard multiplied by the number of shards. For example, for a cluster instance that has 128 shards, in which the memory of a single shard is greater than 1 GB and the maximum bandwidth per shard is 96 MB/s, the overall bandwidth after the direct connection mode is enabled is 12,288 MB/s.

• The bandwidth value applies to the upstream and downstream bandwidths. For example, if the bandwidth of an instance type is 384 Mbps (48 MB/s), the upstream and downstream bandwidths of the instance are both 384 Mbps (48 MB/s).

  Note

  If your instance experiences unexpected or planned traffic spikes, you can adjust the bandwidth of the instance as needed. For more information, see Manually increase instance bandwidth.

• The bandwidth values in the tables are the internal bandwidths of the instance. The Internet bandwidth is determined by the internal bandwidth and is limited by the bandwidth of the connection between the instance and the client. We recommend that you connect to the instance over an internal network to eliminate the impact of the Internet and maximize bandwidth performance.

Maximum connections

• Proxy mode Capped at 500,000 regardless of shard count.

  Instances created before March 1, 2020 have a proxy-mode cap of 200,000.

• Direct connection mode 10,000 connections per shard. Total = 10,000 x (number of shards).

Instance creation methods

You can call the CreateInstanceoperation to create a standalone instance, or create a standalone instance on theSubscription (local disk)tab. For more information about how to create an instance in the console, seeStep 1: Create an instance.

FAQ

My instance's actual bandwidth differs from the spec listed here. Why?

Bandwidth limits for some instance types change after product upgrades. To get the updated bandwidth without extra cost, change the instance configuration and select your current instance type as the target — the system applies the new limits at no additional charge.

Write operations are failing on my cluster instance, but memory isn't full. Why?

Redis distributes keys across shards using a hash algorithm. Large keys can skew this distribution — one shard fills up while others stay empty, causing writes to that shard to fail.

Use the Performance monitoring feature to check memory usage per shard and identify hot shards. See How to view performance metrics such as memory usage of child nodes in a Redis cluster for details.

Does "8proxy" in an instance type code like `redis.logic.sharding.1g.8db.0rodb.8proxy.default` mean there are eight proxy nodes?

No. The 8proxy segment does not represent the number of proxy nodes. Proxy capacity and bandwidth are guaranteed to meet the specs listed in the tables — you don't need to track the actual node count.