Abstract
Computing hardware is constantly evolving and database systems need to adapt to ongoing hardware changes to improve performance. The current hardware trend is heterogeneity, where multiple computing units like CPUs and GPUs are used together in one system. In this paper, we summarize our efforts to use hardware heterogeneity efficiently for query processing. We discuss different approaches of execution and investigate heterogeneous placement in detail by showing, how to automatically determine operator placement decisions according to the given hardware environment and query properties.
Funding statement: This work is funded by the German Research Foundation (DFG) within the Cluster of Excellence “Center for Advancing Electronics Dresden” (Orchestration Path).
About the authors
Tomas Karnagel is currently a Ph.D. student at TU Dresden, working on heterogeneity-aware query optimization. Before starting his Ph.D., he worked with SAP and Intel on in-memory database systems and later he interned at the IBM Almaden Research Center, working on GPU acceleration of database queries.
Database Systems Group, Technische Universität Dresden, Dresden, Germany
Dirk Habich studied Computer Science at the University of Halle-Wittenberg and received his Ph.D. in 2008 from the Technische Universität Dresden, where he still is a member of the scientific staff of Prof. Dr.-Ing. Wolfgang Lehner. His research interest focuses on database support for data mining, in-memory databases, and modern system architectures.
Database Systems Group, Technische Universität Dresden, Dresden, Germany
©2017 Walter de Gruyter Berlin/Boston
