All Products
Document Center

Tair:Common scenarios

Last Updated:Sep 26, 2023

Compared with ApsaraDB for Redis, Tair provides a greater number of popular data structures and is suitable for scenarios that require high concurrency and computing power.

Autonomous driving

Tair provides the time series data structure TS and the geographic information system (GIS) API-compatible data structure GIS that are ideal for autonomous driving scenarios.

  • Scenario 1: Display real-time vehicle trajectory

    Autonomous vehicles need to be monitored and controlled in real time to display real-time trajectory to users. This scenario requires time series queries and has high requirements for real-time reads and writes. TairTS supports downsampling, filtering by attributes, batch queries, and aggregate queries with numeric functions. As such, TairTS can be used to store coordinates of vehicles and display their trajectory.

  • Scenario 2: Configure digital fences

    Level 4 self-driving cars are available only within certain areas. TairTS can be used to configure digital fences so that users can check whether a self-driving car goes out of a digital fence. This use case also extends to shared bikes and electric bikes.

Live streaming

Live streaming has been gaining popularity and is widely used in fields such as e-commerce, meetings, gaming, and sports events. Tair provides enterprise-level high concurrency that serves live streaming better.

  • Scenario 1: Store bullet screens (scrolling comments)

    Bullet screens are essentially real-time comments that feature high read and write queries per second (QPS). Tair can be used to cache scrolling comments with time attributes. This way, your application can obtain the comments within a time range and display them on clients.

  • Scenario 2: Live streaming playbacks

    To respond to new needs, live streaming allows users to view playbacks while the live streaming is ongoing. Tair can be used to store time series indexes, generate audio and video index files, and replay videos by using the HTTP Live Streaming (HLS) protocol.

  • Scenario 3: Store live streaming message queues

    Live streaming involves multi-terminal communications, and therefore the live streaming server must cope with multi-terminal data consistency. Tair can be used to store information from multiple terminals, read offset information, and ensure multi-terminal data consistency.


High concurrency is typical of the gaming industry, especially for new game launches and gaming events. Tair provides scalability and a variety of data structures to build gaming platforms with high availability and low costs.

  • Scenario 1: Cache data of gaming events

    High concurrency spikes are common during gaming events. In this case, Tair allows you to scale your instances to handle traffic spikes and prevent cache breakdown of the backend database.

  • Scenario 2: Display leaderboards

    The leaderboard is a popular feature for gaming applications. The time to live (TTL) feature of exZset provided by Tair can be used to implement the overall leaderboard and user-level leaderboards. This improves user experience.