This topic describes the specifications of Tair (Enterprise Edition) ESSD/SSD-based instances. The specifications include the memory capacity, maximum number of connections, and maximum bandwidth.
Instance specifications
The following tables list the specifications of individual shards.
The overall performance of a standard instance is determined by the specifications of the shards in the instance.
The overall performance of a cluster instance can be estimated by multiplying the performance of a single shard by the total number of shards in the instance.
NoteStarting May 8, 2024, the maximum number of connections to each shard of an ESSD/SSD-based instance is increased to 60,000 or more. This increase applies when you create an instance or upgrade the specifications of an existing instance. For specific information, see the following table.
Starting March 14, 2025, Tair (Enterprise Edition) upgrades ESSD-based instances of the 32C - 128GB specification or higher (tair.essd.standard.8xlarge, tair.essd.standard.13xlarge, and tair.essd.standard.16xlarge) and SSD-based instances of the 32C - 256GB specification or higher (tair.localssd.c1m8.8xlarge and tair.localssd.c1m8.with.proxy.8xlarge). When you create instances or upgrade the specifications of existing instances, the maximum number of connections per shard is increased to 100,000.
Instances that can be created in different regions and zones are subject to actual inventory.
Shard specifications of ESSD-based instances
ESSD-based instances support only the standard architecture.
Specification type | InstanceClass value (used in API operations) | CPU cores | Memory (GB) | Storage performance level and capacity range (GB) | Maximum connections | Bandwidth |
4C - 16GB | tair.essd.standard.xlarge | 4 | 16 | PL1: 60 to 250 | 60,000 | 1,500 Mbps (187.5 MB/s) |
8C - 32GB | tair.essd.standard.2xlarge | 8 | 32 | PL1: 60 to 500 PL2: 470 to 1,000 | 60,000 | 2,000 Mbps (250 MB/s) |
16C - 64GB | tair.essd.standard.4xlarge | 16 | 64 | PL1: 60 to 1,000 PL2: 470 to 2,000 PL3: 1,270 to 4,000 | 60,000 | 3,000 Mbps (375 MB/s) |
32C - 128GB | tair.essd.standard.8xlarge | 32 | 128 | PL2: 470 to 4,000 PL3: 1,270 to 8,000 | 100,000 | 5,000 Mbps (625 MB/s) |
52C - 192GB | tair.essd.standard.13xlarge | 52 | 192 | PL2: 470 to 6,000 PL3: 1,270 to 12,000 | 100,000 | 8,000 Mbps (1,000 MB/s) |
64C - 256GB | tair.essd.standard.16xlarge | 64 | 256 | PL2: 470 to 8,000 PL3: 1,270 to 16,000 | 100,000 | 8,000 Mbps (1,000 MB/s) |
Enterprise SSDs at performance level 3 (PL3 ESSDs) perform better than PL2 and PL1 ESSDs. For more information, see ESSDs.
Shard specifications of SSD-based standard instances and cluster instances in direct connection mode
The following specifications are applicable to SSD-based standard instances and cluster instances in direct connection mode.
Specification type | InstanceClass value (used in API operations) | CPU cores | Memory (GB) | Storage performance level and capacity range (GB) | Maximum connections | Bandwidth |
4C-32GB | tair.localssd.c1m8.xlarge | 4 | 32 | 640 | 60,000 | 1,500 Mbps (187.5 MB/s) |
8C-32GB | tair.localssd.c1m4.2xlarge | 8 | 32 | 320 | 60,000 | 3,000 Mbps (375 MB/s) |
8C-64GB | tair.localssd.c1m8.2xlarge | 8 | 64 | 1,280 | 60,000 | 2,500 Mbps (312.5 MB/s) |
16C-64GB | tair.localssd.c1m4.4xlarge | 16 | 64 | 640 | 60,000 | 5,000 Mbps (625 MB/s) |
32C-128GB | tair.localssd.c1m4.8xlarge | 32 | 128 | 1,280 | 100,000 | 12,000 Mbps (1,500 MB/s) |
Shard specifications of SSD-based cluster instances in proxy mode
The following specifications are applicable to SSD-based cluster instances in proxy mode.
Specification type | InstanceClass value (used in API operations) | CPU cores | Memory (GB) | Storage performance level and capacity range (GB) | Maximum connections | Bandwidth |
4C-32GB | tair.localssd.c1m8.with.proxy.xlarge | 4 | 32 | 640 | 60,000 | 1,500 Mbps (187.5 MB/s) |
8C-32GB | tair.localssd.c1m4.with.proxy.2xlarge | 8 | 32 | 320 | 60,000 | 3,000 Mbps (375 MB/s) |
8C-64GB | tair.localssd.c1m8.with.proxy.2xlarge | 8 | 64 | 1,280 | 60,000 | 2,500 Mbps (312.5 MB/s) |
16C-64GB | tair.localssd.c1m4.with.proxy.4xlarge | 16 | 64 | 640 | 60,000 | 5,000 Mbps (625 MB/s) |
32C-128GB | tair.localssd.c1m4.with.proxy.8xlarge | 32 | 128 | 1,280 | 60,000 | 12,000 Mbps (1,500 MB/s) |
Performance description
Architecture | Performance description |
The overall performance of standard instances is the same as that described in the preceding instance specification tables. | |
Overall performance of an instance = Performance of a shard × Number of shards. Note When a cluster instance that runs in proxy mode is accessed, the maximum bandwidth is 16,384 Mbps (2,048 MB/s), the maximum total QPS is 10,000,000, and the maximum number of connections is 500,000. For example, an instance that contains four shards of the tair.localssd.c1m4.with.proxy.2xlarge type has the following specifications:
In this case, the instance provides the following performance:
|
Notes about CPU cores
To ensure service stability, the system reserves a CPU core to process background tasks. In a cluster instance or a read/write splitting instance, the system reserves a CPU core for each shard or read replica to process background tasks.
Bandwidth description
Each bandwidth value in the preceding tables represents the maximum bandwidth of a single shard. The total bandwidth of a cluster or read/write splitting instance is the combined bandwidth of all shards in the instance.
The maximum bandwidth of a cluster instance in proxy mode or a read/write splitting instance is 16,384 Mbps (2,048 MB/s). After the upper limit is reached, the bandwidth cannot be increased even if you add more shards to the instance. Cluster instances in direct connection mode have no limits on the overall bandwidth.
The bandwidth applies to the upstream and downstream bandwidths. For example, if the overall bandwidth of an instance is 768 Mbps (96 MB/s), the upstream and downstream bandwidths of the instance are both 768 Mbps (96 MB/s).
The bandwidth limits of Tair and Redis Open-Source Edition 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.
For information about the FAQ about bandwidth, see FAQ about bandwidth.
Calculation rules for connections
Item | Description |
Maximum number of connections |
|