CAREER: Bridging Databases and Computer Architecture: Optimizing DBMS for Deep Memory Hierarchies

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

This research project will develop techniques to make database systems cache-resident. The approach will develop three classes of techniques and they are (i) static data placement for deep memory hierarchies, (ii) cache buffer management and (iii) addressing coherence overload on multiprocessor systems. The research will also develop optimization techniques and investigate data layout alternatives. The career development plan will bridge database systems and computer architecture education programs and also promote interactions between computer architecture and database systems students and researchers.