What type of latency does the DynamoDB Accelerator (DAX) achieve for cached reads and writes?

DynamoDB Accelerator (DAX) is designed to provide in-memory caching for DynamoDB, significantly reducing the latency associated with read and write operations. By storing copies of frequently accessed data in memory, DAX optimizes access speeds to achieve performance enhancements.

The correct answer indicates that DAX can achieve latency in the microsecond range for cached reads and writes. This is critical for applications requiring rapid response times, as it allows for nearly instantaneous access to cached data compared to standard operations that may take milliseconds.

Other choices reflect longer latency ranges, which do not apply to DAX's capabilities. By leveraging in-memory storage and efficient data retrieval techniques, DAX enables applications to scale efficiently while maintaining high performance, especially beneficial for workloads with heavy read demand.