What is Tair? - - Alibaba Cloud Documentation Center

Tair is a cloud-native in-memory database service that is developed by Alibaba Cloud. Tair is compatible with open source Redis and provides a variety of data models and enterprise-class features to support real-time online applications. Tair also introduces persistent memory-optimized instances that are based on the new non-volatile memory (NVM) storage medium. Compared with DRAM-based instances, persistent memory-optimized instances offer cost savings of over 30% while ensuring data persistence and providing almost the same performance. Tair has been widely used by customers in sectors such as public service, finance, manufacturing, healthcare, and pan-Internet to meet their high-speed query and computing requirements.

Why Tair?

A variety of series types of instances

Tair introduces multiple series types of instances based on storage mediums such as dynamic random-access memory (DRAM), non-volatile memory (NVM), and Enterprise SSDs (ESSDs) to meet your requirements for low-latency access, data persistence, and reduced overall costs. Tair provides you with higher performance, more data structures, and more flexible storage methods. This helps you meet business requirements in different scenarios. For more information, see DRAM-based instances, Persistent memory-optimized instances, and ESSD/SSD-based instances.

Enterprise-class availability

Tair provides enterprise-class features such as data flashback and hotkey optimization to ensure data reliability and instance stability. Tair also supports imperceptible scaling for cluster instances. This scaling solution prevents data loss that often occurs in frequently-used scaling options.

A wide range of data structures

Tair provides a variety of data structures, such as TairString, TairHash, TairGIS, TairCpc, and TairBloom, to help you reduce costs and innovate.

Note

For more information, see Benefits and Common scenarios.

Storage types

The rapid development of the Internet makes business scenarios more diverse and complex. To meet the changing requirements, Tair, as a high-availability and high-performance NoSQL database service, provides multiple series types of instances based on storage mediums such as DRAM, NVM, and ESSDs. These series types are developed to meet your requirements for low-latency access, data persistence, and reduced overall costs, and provide higher performance, more data structures, and more flexible storage methods. This helps you meet business requirements in different scenarios.

Storage type	Description
DRAM-based instance	Provides ultra-high performance. DRAM-based instances use the multi-threading model to provide read and write performance three times that of native Redis databases or ApsaraDB for Redis Community Edition instances that have the same specifications. For more information, see Performance whitepaper of Redis 5.0-compatible Tair DRAM-based instances and Performance whitepaper of Redis 6.0-compatible Tair DRAM-based instances. Provides a variety of self-developed enhanced data modules. These modules include exString (including commands that enhance Redis string functionality), exHash, exZset, GIS, Bloom, Doc, TS, Cpc, Roaring, Search, and Vector. These modules allow you to spend less time writing code, improve business performance, and focus on innovation. Provides enterprise-class features, such as data flashback based on point-in-time recovery (PITR), proxy query cache, and Global Distributed Cache. Supports enterprise-class encryption methods, such as Transport Layer Security (TLS) encryption and transparent data encryption (TDE). For more information, see Enable TLS encryption and Enable TDE.
Persistent memory-optimized instance	Offers ultra-high cost-effectiveness. The price of persistent memory-optimized instances is 30% lower than that of ApsaraDB for Redis Community Edition instances that have the same capacity. The performance of persistent memory-optimized instances reaches 90% of that of ApsaraDB for Redis Community Edition instances. For more information, see Performance white paper of persistent memory-optimized instances. Supports enhanced data modules. These modules include exString (including commands that enhance Redis string functionality), exHash, and Cpc. Prevents data loss caused by power failure. Persistent memory-optimized instances implement persistence for each command. The system returns a success response for each write operation only after the data is persistently stored. You can use persistent memory-optimized instances as in-memory databases instead of caches.
ESSD/SSD-based instance	Reduced costs: ESSD/SSD-based instances reduce up to 85% of costs compared with ApsaraDB for Redis Community Edition instances. High performance: An ESSD/SSD-based instance delivers approximately 60% of the performance of an ApsaraDB for Redis Community Edition instance. For more information, see Performance whitepaper of ESSD-based instances. Storage in disks: ESSD/SSD-based instances store data in ESSDs or SSDs to ensure high data reliability. The capacity of an ESSD/SSD-based instance can reach hundreds of terabytes. Data distribution: ESSD/SSD-based instances use the Alibaba Cloud TairDB storage engine and combine disks and memory to provide an optimal balance between data persistence and quick access to data. High compatibility: ESSD/SSD-based instances are compatible with most data structures and commands of ApsaraDB for Redis Community Edition instances that run Redis 6.0.

Instance architectures

Tair offers multiple instance deployment architectures to support a variety of scenarios.

Architecture	Description
Standard architecture	Master-replica deployment model: An instance consists of a master node and a replica node. The master node handles your daily workloads, and the replica node stays in hot standby mode to ensure high availability. If the master node fails, the high-availability system switches the workloads to the replica node within seconds after the failure occurs. This mechanism ensures the stability of your workloads. Standalone deployment model: An instance contains only a single data node that cannot ensure high availability. Standalone standard instances are suitable for cache-only scenarios that do not require high data reliability. You can scale up the data node to meet high queries per second (QPS) requirements in a cost-efficient manner.
Cluster architecture	Distributed deployment model: Each data shard works as a master-replica high-availability node and provides automatic disaster recovery and failover. Standalone deployment model: Each data shard contains only a single data node that cannot ensure high availability. Standalone cluster instances are suitable for cache-only or high QPS scenarios. Note The cluster architecture supports the following connection modes: Proxy mode: manages connections to your instance in an intelligent manner to reduce the costs of application development. Direct connection mode: allows a client to bypass proxy nodes and directly access backend data shards to reduce the network overhead and service response time. This mode is suitable for latency-sensitive business scenarios.
Read/write splitting architecture	Read/write splitting instances use the master-replica architecture to provide high availability. Read replicas are attached to the master node for data replication and linear scaling of read performance. Read replicas can alleviate performance issues caused by hotkeys. Read/write splitting instances are suitable for business scenarios that feature high read/write ratios. The read/write splitting architecture provides two types of instances: Cloud-native instance: Read replicas synchronize data from the master node by using star replication. You can specify up to nine read replicas. This type is suitable for business scenarios that feature ultra-high read/write ratios. Classic instance (retired): Read replicas use chained replication. You can specify one, three, or five read replicas.

Instance types

Tair provides multiple architectures and series types of instances. For more information, see Overview.

Purchase method

Create a Tair instance

Millstones of Tair

In 2004, Taobao started using caching technologies to support its business operations. The first caching technology that was put into use is frontend page caching. This technology uses Edge Side Includes (ESI) to identify web content segments that can or cannot be accelerated. Frontend page caching reduces the number of times that entire pages are fetched from the server.

The rapid growth of customer traffic on Taobao exerted more pressure on databases. Backend caching emerged in response to this situation. Backend caching has evolved over many iterations from Tbstore to Taobao Database Manager (TDBM). Tbstore provides services such as persisting Taobao details and verification codes. TDBM was initially used in Taobao User Center. In 2009, the large-scale and high-speed storage service Tair was released. This solution was developed based on the success of previous systems and technical experience.

Tair is one of the most popular systems of Alibaba Group and has provided core access acceleration during Double 11 for many years. Tair can handle hundreds of millions of calls per second.

Date	Event
July 2024	Redis 7.0-compatible DRAM-based instances were released with comparable performance to Redis 6.0-compatible DRAM-based instances.
October 2022	Redis 6.0-compatible DRAM-based instances were available with a 20% performance improvement than Redis 5.0-compatible DRAM-based instances that have the same specifications.
July 2021	The following new series types were released: Persistent memory-optimized instances use persistent memory to provide large-capacity in-memory databases that are compatible with open source Redis. Persistent memory-optimized instances persist each operation without using disks and cost up to 30% less than ApsaraDB for Redis Community Edition instances while delivering almost identical throughput and latency. This helps improve the reliability of business data. ESSD/SSD-based instances are developed based on ESSDs or SSDs and are compatible with core data structures and APIs of open source Redis. These instances can provide large-capacity, low-cost, and persistent database services. Tair will focus on core cloud-native capabilities, such as hardware-software integration, intelligent data distribution, and integration of data storage and computing.
November 2019	Tair 3.0 or ApsaraDB for Redis Enhanced Edition (Tair) was released. Redis 5.0-compatible DRAM-based instances use the multi-threading model. These instances integrate multiple Tair data structures to provide high-performance, high-compatibility database services that come with enterprise-class features.
April 2019	The Tair team was recognized as one of the top three contributors in the Redis community and delivered a public speech at RedisConf 2019.
August 2018	Tair took the lead in introducing hybrid-storage instances in China to separate cold data from hot data for the first time and reduce costs.
November 2017	Tair dynamic hashing provided support for the 2017 Double 11 event and was able to resolve several critical cache issues within the industry.
April 2017	Tair 2.0 was released and started to provide services to the new business units AMAP and Youku. ApsaraDB for Memcache (OCS) was upgraded to ApsaraDB for Redis.
August 2016	Tair Smart O&M Platform was released. The release of this platform contributed to the breakthrough of hundreds of billions of sales during the 2016 Double 11 event.
March 2015	The Tair team released ApsaraDB for Redis. This release helped Tair enter the cloud era.
May 2014	The Tair team released OCS, a cloud-based caching service. After OCS was released, OCS became a basic service of Alibaba Cloud and started to provide services for OCS users.
April 2013	The Fastdump service was released. It can significantly reduce import time and access latency. Tair was implemented on a large scale in Alimama.
October 2012	The Redis Database (RDB) cache engine was released with APIs that are similar to those used in open source Redis. This engine supports more flexible and complex data structures.
June 2011	The Local Database (LDB) persistence engine was released to meet key-value storage requirements in the Internet industry.
November 2009	Tair was used to support the high-volume concurrent traffic during the Double 11 event for the first time.
April 2009	Tair 1.0 was released and used in business segments such as the Taobao core system, Memory-Mapped Database (MDB) cache, and User Center.