Tair provides three types of instances: DRAM-based instances, persistent memory-optimized instances, and ESSD-based instances. You can perform the procedure in this topic to create a Tair instance that meets your business requirements.
An Alibaba Cloud account is created. For more information, see Sign up with Alibaba Cloud.
If you want to create a pay-as-you-go instance, make sure that you have sufficient balance within your account.
Go to the buy page.
Configure the parameters that are described in the following table.
The product type of the instance. By default, Tair (Cloud Disk) is selected.
The billing method of the instance. The parameter is automatically set to Subscription.
Memory: uses a multi-thread model, delivers three times the performance of a Community Edition Redis instance with the same specifications, and provides a variety of enhanced data structures to simplify development. For more information, see DRAM-based instances.
Persistent Memory: allows you to store and query data in persistent memory. This type provides command-level persistence capabilities. It is suitable for scenarios that require high performance and data consistency. For more information, see Persistent memory-optimized instances.
ESSD: stores data in enhanced SSDs (ESSDs), provides command-level persistence capabilities and massive storage capacity. It is suitable for scenarios that require moderate performance and high cost-effectiveness. For more information, see ESSD-based instances.
The region in which you want to create the instance. The region cannot be changed after the instance is created.
To maximize access speed, we recommend that you select a region that is close to the geographic location of your users.
To enable connections over an internal network, make sure that the Tair instance is deployed in the same region as Elastic Compute Service (ECS) instances that need to access the Tair instance. Otherwise, the Tair instance cannot communicate with the ECS instances over an internal network to achieve optimal performance.
Single-zone: The master and replica nodes are deployed in the same zone. This deployment type can provide single-data center disaster recovery.
Dual-zone: The master node and replica node are separately deployed in the specified primary zone and secondary zone of the same region to form a zone-disaster recovery architecture.
The zone in which you want to create the instance. Each region has multiple isolated locations that are known as zones. Each zone has its own independent power supply and network. When an ECS instance and a Tair instance are deployed in the same zone, they can communicate over an internal network to minimize the network latency.Note
If you set the Deployment Type parameter to Dual-zone, you must also set the Secondary Zones parameter to a zone that is different from the zone specified by the Primary Zone parameter. The primary and secondary zones together form a master-replica zone-disaster recovery architecture.
The network type of the instance. This parameter can be set only to VPC. A virtual private cloud (VPC) is an isolated network that provides higher security and better performance than the traditional classic network.Important
Make sure that the Tair instance is deployed in the same VPC as the ECS instances or ApsaraDB RDS instances that need to access the Tair instance. ECS instances located in other VPCs can access the Tair instance only over the Internet.
The VPC in which you want to create the instance. If you do not have a VPC, create one first. For more information, see Create and manage a VPC.
The vSwitch that you want to connect to the instance in the VPC. If no vSwitches are available in the VPC in the current zone, create a vSwitch. For more information, see Create and manage a vSwitch.
The parameter is automatically set to High Availability.
The engine version of the instance that you want to create.
DRAM-based instances support Redis 5.0 and Redis 6.0.
Persistent memory-optimized instances support Redis 6.0.
ESSD-based instances support Redis 4.0.
Select whether to enable Read/Write Splitting.
If you disable Read/Write Splitting, the instance uses the standard master-replica architecture.
If you enable Read/Write Splitting, the instance uses the read/write splitting architecture. In this case, you must specify the Replicas parameter. Valid values: 2 to 6. A value in this range includes a master node. Up to five read replicas are supported.
Cluster: uses the sharded cluster architecture. For more information, see Cluster architecture.
If you select Cluster, you must perform the following operations:
Specify Connection Mode:
Proxy (default): uses proxy nodes to connect to the Tair instance, enable architecture changes, route commands, balance loads, and perform failovers. The connection mode can aggregate connections and enhance read performance. In addition, this mode is easy to use and simplifies your business system. For more information, see Features of proxy nodes.
Direct Connection: allows you to directly connect to the Tair instance in the same manner as you connect to open source Redis.
After the instance is created, you cannot change the connection mode for the instance. You can select only a single connection mode for your instance. We recommend that you evaluate your use scenario in advance and select the connection mode that suits your business requirements.
Specify Shards. The instance performance increases with the number of shards. You can calculate the instance capacity by using the following formula: Instance capacity = Number of shards × Shard specifications. You can select 2 to 64 shards when you create an instance. After you create the instance, you can add more shards to the instance. For more information, see Adjust the number of data shards for a Tair instance with cloud disks.
ESSD-based instances do not support the sharded cluster architecture.
The Connection Mode parameter is available only when the Architecture parameter is set to Cluster.
After the instance is created, you cannot change the connection mode for the instance. You can select only a single connection mode for your instance. We recommend that you evaluate your use scenario in advance and select the connection mode that best suits your business requirements.
Specifies whether to enable read/write splitting.
The specifications of the instance. Each option contains a group of configurations, such as the memory capacity, maximum number of concurrent connections to the instance, and maximum bandwidth. For more information, see Instance specifications.Note
Database metadata is generated when a Tair instance is created. The size of the metadata on each shard of a cluster instance can range from 30 MB to 50 MB. The total size of the metadata on a cluster instance equals the total size of metadata on all shards of the cluster instance.
If you enable Read/Write Splitting, the Nodes parameter can be specified. Otherwise, this parameter is automatically set to 2. Each shard has a master node and a replica node to ensure high availability (HA).
Read Replica Quantity
The number of read replicas. The value of this parameter varies based on the value of the Nodes parameter.
The number of Shards. The instance performance increases with the number of shards. You can calculate the instance capacity by using the following formula: Instance capacity = Number of shards × Shard specifications. You can select 2 to 64 shards when you create an instance. After you create the instance, you can add more shards to the instance. For more information, see Adjust the number of data shards for a Tair instance with cloud disks.
The parameter is automatically set to ESSD PL1.Note
This parameter is available only when you set the Storage Type parameter to ESSD.
The storage capacity of the instance. After the instance is created, you can only increase the storage capacity.Note
This parameter is available only when you set the Storage Type parameter to ESSD.
Set Later: sets the password after the instance is created. For more information, see Change or reset the password.
Set Now: sets the password immediately.
The password must be 8 to 32 characters in length.
The password must contain at least three of the following character types: uppercase letters, lowercase letters, digits, and special characters.
These special characters include
! @ # $ % ^ & * ( ) _ + - =.
The name of the instance. Specify a name that is easy for you to identify.
The subscription duration of the instance.
The resource group to which you want to assign the instance. For more information, see What is Resource Management?
Click Buy Now.
On the Confirm Order page, read and select the Terms of Service and complete the payment as prompted.
After you complete the payment, wait 1 to 5 minutes. Go to the Tair console and select the region in which the Tair instance is deployed. The instance is displayed on the Instances page.
Q: How long does it take to create an instance?
A: The amount of time that is required to create an instance depends on the number of shards in the instance. A larger number of shards require more resources. As a result, it takes longer to create the instance. For example, it can take 2 to 3 minutes to create a standard master-replica instance, 10 to 15 minutes to create a 128-shard cluster master-replica instance, and 20 to 40 minutes to create a 256-shard cluster master-replica instance.Note
A standard instance is regarded as a single-shard instance.
The master node of a read/write splitting instance is regarded as a shard that processes both read and write requests. Each read replica of a read/write splitting instance is regarded as a shard that processes only read requests.
Q: Why am I unable to find the created instance?
A: The following table describes the possible causes and solutions.
The region that you selected in the Tair console is not the region in which the instance is deployed.
Log on to the Tair console and select the region in which the instance resides.
The instance list in the Tair console is not updated or is updated before the instance is created.
Wait several minutes and then update the instance list to check whether the Tair instance is added to the list.
The instance is not created due to insufficient resources.
The system may fail to create the instance due to insufficient resources. In this case, your payment is refunded. You can check your refund on the Orders page.
Creates a Tair instance.