User-defined functions such as currency conversion and factory calendar are important ingredients in many business applications. Since currency conversion and factory calendar are expensive userdefined functions, optimizing these functions is essential to high performance business applications. We optimize scalar user-defined functions by caching function call results. In this paper we investigate which method for function result caching is best in the context of in-memory column-store database systems. Experiments show that our method, which implements a function result cache as an array, combined with SAP HANA in-memory column store provides the high performance required by real-time global business applications.
CITATION STYLE
Ryu, C., Lee, S., Kim, K., Park, K., Kwon, Y., Cha, S. K., … Muench, S. (2017). Optimizing scalar user-defined functions in in-memory column-store database systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10178 LNCS, pp. 568–580). Springer Verlag. https://doi.org/10.1007/978-3-319-55699-4_35
Mendeley helps you to discover research relevant for your work.