Datenschutzerklärung|Data Privacy
Impressum

07.06.2012
A. Borusan

29.06.2012, 11 Uhr s.t. TU Berlin, EN building, seminar room EN 719 (7th floor), Einsteinufer 17, 10587 Berlin: "Query Compilation and Execution in HyPer" (Prof. Thomas Neumann, TU M√ľnchen)

The traditional way to execute queries is the so called iterator model, where
algebraic operators produce tuple streams on demand. While being very flexible,
the iterator model is not well suited for main-memory database systems, as it
is very cache-unfriendly and causes a significant overhead. Data-centric
execution models that blur operator boundaries to maximize locality reach a
significant better performance on modern CPUs. The HyPer systems compiles both
regular SQL queries and more complex pre-canned transactions into such
execution models using LLVM for machine-code generation.