A Multi-Granularity Annotation Management and Multi-Scale Provenance Compression Technique for Data Streams

In this paper we propose a new technique for multi-granularity provenance annotation management and multi-scale provenance compression applied to data streams. First of all, we propose a new mgp-SQL syntax system based on some studies. The construction and propagation of data provenance annotations in three different granularities have been described in this paper. Provenance of multiple granularities can be propagated in the process of querying. Then, we introduce a compression method using DWT (Discreet Wavelet Transform) for data provenance carried by data streams. We have proposed a novel multi-scale conception for efficient storage which allows users to adjust compression ratio in the process of continuous querying. Thus, provenance annotations of a proper scale will be constructed and stored. Finally, we combine both techniques and present the algorithm.