Abstract
In the era of social media and big data, many organizations and countries are devoting considerable effort and money to reduce energy consumption. Despite that, current research mainly focuses on improving performance without taking into account energy consumption. Recently, great importance has been attached to finding a good compromise between energy efficiency and performance in data warehouse (DW) applications. For a given DW, multiple logical schemes may exist due to the presence of dependencies and hierarchies among the attributes. In this respect, it has been shown that varying the logical schema has an impact on energy saving. In this paper, we introduce a new approach for efficient exploration of the different logical schemes of a DW. To do so, we prune the search space by relying on anti-monotonicity based constraint to swiftly find the most energy-efficient logical schema. The carried out experiments show the sharp impact of the logical design on energy saving.
