百度智能云

All Product Document

          Simple Cache Service SCS

          Business Scenarios

          Data Application Scenarios

          Cache

          Equip the persistent storage to ease the pressure on back-end storage services. As users' tolerance for application delay is getting lower increasingly, but the disk-based random data access performance has not increased revolutionarily, the data access performance has become a shortcoming in product performance. The distributed cache service can be used to effectively put the access data in the cache. The data in the cache are consistent with those in the back-end disk storage system through a synchronous or asynchronous mechanism. Most data access requests can directly hit the cache to achieve a response time being orders of magnitude lower than the disk access. When the data access request does not hit the cache, you need to access the disk storage system and update the cache.

          Key Value Storage

          You can use the distributed cache separately in services in need of fast response. As the cache service becomes more reliable and more convenient, when the data of certain products does not require the strong support and consistency, all user data can be stored in the cache by weakening the support for the transactions and data consistency. The key value storage can provide abundant data storage modes, which is more business-friendly in the Internet scenario, and can achieve higher scalability, performance and reliability.

          Industry Application Scenarios

          Applications in the feed stream industry

          There are a lot of Redis applications in the feed stream industry. For example, Redis can be used to filter the user's delivery history to ensure the user's feed stream recommendations are not repeated. Various likes and comments in the feed stream can also be stored in Redis to achieve quick interaction between users and feed streams.

          Applications in the Financial Industry

          Extensive risk control check items are stored in Redis in the risk control scenes of the financial industry can achieve user images in seconds; during the card binding process, the user's identity information is stored in Redis to achieve good user feedback.

          Applications in AI Industry

          In various AI session services, the session between the robots and the users is often stored in Redis cache, and AI application can implement the multiple rounds of interaction by processing sessions for multiple times to achieve a higher level of AI conversation.

          Access Count

          Redis’s atomic counting, INCR, INCRBY and other commands can be used to implement the most typical Redis applications, support the expiration configurations and policies and batch operation, which can ensure the high performance of the system.

          Profile

          The built-in data structures such as Hash or List can be used to store personal profiles. This type of data has a large amount of reads/writes, but each piece of data is not large in its length, and specific fields may be often updated.

          Leaderboard

          It is related to the services such as ranking or zipper with the sorting requirement and supports real-time updates to avoid the performance loss resulted from placement of the sorting logic in the business code.

          Cache

          Compared with other Cache services, Redis has unique multi-data structure support, which is efficient and fast. Redis engine internally supports multiple data elimination algorithms.

          Buffer Queue

          You can use List to implement simple data queue caching to provide a high-performance support for the system.

          Distributed Lock DLM

          For distributed lock businesses, Redis can provide the high-performance distributed lock services, such as: spike, global increment ID, etc.

          Previous
          Instance Specification
          Next
          Instructions for Use