The use of a single thread to operate on memory data is to avoid competition for shared resources and the resource consumption caused by multiple threads waiting simultaneously. If we consider using a separate thread for each data structure for operations, it would utilize multiple cores, while maintaining the same behavior as single-threaded when operating on the same object. However, this would increase the CPU's multi-core capabilities, thereby improving throughput, without affecting the read/write operations of other data structures. But this is just my simple speculation, and I'm not sure if the complexity of implementing this approach would be high in reality.