Effective Skyline Cardinality Estimation on Data Streams

摘要: In order to incorporate the skyline operator into data stream engine, we need address problem of cardinality estimation, which is very important for extending query optimizer's cost model accommodate queries. this paper, propose robust approaches estimating over sliding windows in environment. We first design an approach estimate uniformly distributed data, and then extend support arbitrarily data. Our allow arbitrary distribution, hence can be applied model. To online manner, live elements window are sketched using Spectral Bloom Filters efficiently effectively capture information essential windows. Extensive experimental study demonstrates that our significantly outperform previous approaches.