This research presents a procedure and structure for efficient distributed computation and massive data query in online analyzing and processing. The approach utilizes a cluster structure to enable distributed pre-computation and query operations on data cubes. The key innovation lies in the partitioning of a large-capacity dataset into multiple blocks distributed across nodes using the MapReduce framework. Each node performs local closed cube computation through Map tasks, and parallel query operations are executed on different nodes to retrieve multiple measuring values from local closed cubes. The measuring values are then merged using Reduce tasks. This procedure offers several advantages, including simplified and effective pre-computation and query processes for large-capacity data in online analysis, reduced storage space requirements for data cubes, and rapid response times for user queries.

Distributed computation, massive data query, online analyzing and processing, Data cubes, MapReduce, Cluster structure, Pre-computation, User query.