Abstract
Workflow records are naturally depicted using a graph model in which service points are denoted as nodes in the graph and edges portray the flow of processing. In this paper we consider the problem of enabling aggregation queries over large-scale graph datasets consisting of millions of workflow records. We discuss how to decompose complex, ad-hoc aggregations on graph workflow records into smaller, independent computations via proper query rewriting. Our framework allows reuse of precomputed materialized query views during query evaluation and, thus, enables view selection decisions that are of immense value in optimizing heavy analytical workloads. © 2012 Springer-Verlag.
Cite
CITATION STYLE
Bleco, D., & Kotidis, Y. (2012). A framework for enabling query rewrites when analyzing workflow records. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7338 LNCS, pp. 587–590). https://doi.org/10.1007/978-3-642-31235-9_40
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
