Serving datacube tuples from main memory
Kenneth Andrew Ross, Kazi A. Zaman · 2002
Existing datacube precomputation schemes materialize selected datacube tuples on disk, choosing the most beneficial cuboids (i.e., combinations of dimensions) to materialize given a space limit. However in the context of a data-warehouse receiving frequent "append" updates to the database, the cost of keeping these disk-resident cuboids up-to-date can be high. In this paper we propose a main memory based framework which provides rapid response to queries and requires considerably less maintenance cost than a disk based scheme in an append-only environment. For a given datacube query, we first look among a set of previously materialized tuples for a direct answer. If not found, we use a hash based scheme reminiscent of partial match retrieval to rapidly compute the answer to the query from the finest-level data stored in a special in-memory data structure. Our approach is limited to the important class of applications in which the finest granularity tuples of the datacube fit in main memory. We present analytical and experimental results demonstrating the benefits of our approach.