Lineage retrieval for scientific data processing: A survey

Scientific research relies as much on the dissemination and exchange of data sets as on the publication of conclusions. Accurately tracking the lineage (origin and subsequent processing history) of scientific data sets is thus imperative for the complete documentation of scientific work. Researchers are effectively prevented from determining, preserving, or providing the lineage of the computational data products they use and create, however, because of the lack of a definitive model for lineage retrieval and a poor fit between current data management tools and scientific software. Based on a comprehensive survey of lineage research and previous prototypes, we present a metamodel to help identify and assess the basic components of systems that provide lineage retrieval for scientific data products.