Approximate Calculation of Window Aggregate Functions via Global Random Sample

• DOI: 10.1007/s41019-018-0060-x
• Publication Date: 2018-03-15T06:09:26Z
• Authors: Guangxuan Song; Wenwen Qu; Xiaojie Liu; Xiaoling Wang
• Keywords: Benchmark (surveying); Implementation; Sample (material)

Window functions have been a part of the SQL standard since 2003 and studied extensively during past decade. They are widely used in data analysis; almost all current mainstream commercial databases support window functions. However, recent years size datasets is growing steeply; existing function implementations not efficient enough. Recently, some sampling-based algorithms (e.g., online aggregation) proposed to deal with large complex relational databases, which offer us flexible trade-off between accuracy efficiency. few sampling techniques has considered for databases. In this paper, we extend our previous work (Song et al. Asia-Pacific web web-age information management joint conference on big data, Springer, pp 229–244, 2017) two new algorithms: range-based global algorithm row-labeled algorithm. The use rather than local more other algorithms. And find out performed baseline method over TPC-H benchmark dataset.