Distributed query refers to the process of retrieving large datasets in a distributed environment. Compared to centralized storage structures, the relational tables in distributed queries are fragmented or replicated and stored on different physical nodes. As shown in Fig. 3.1, during the implementation of a distributed query, the system needs to extract relevant data from various physical storage nodes and transmit the data over the network to the computation nodes for join processing, thus introducing additional network communication overhead and increasing the query processing response time.