Adaptive Access Path Selection for Hardware-Accelerated DRAM Loads

  • Markus DreselerEmail author
  • Timo Gasda
  • Jan Kossmann
  • Matthias Uflacker
  • Hasso Plattner
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10837)


For modern main memory database systems, the memory bus is the main bottleneck. Specialized hardware components of large NUMA systems, such as HPE’s GRU, make it possible to offload memory transfers. In some cases, this improves the throughput by 30%, but other scenarios suffer from reduced performance. We show which factors influence this tradeoff. Based on our experiments, we present an adaptive prediction model that supports the DBMS in deciding whether to utilize these components. In addition, we evaluate non-coherent memory access as an additional access method and discuss its benefits and shortcomings.



We thank Martin Boissier and Rainer Schlosser for their helpful input on the estimation model.


  1. 1.
    Abadi, D.J., Madden, S., Ferreira, M.: Integrating compression and execution in column-oriented database systems. In: ACM SIGMOD International Conference on Management of Data (2006)Google Scholar
  2. 2.
    Boncz, P.A., Manegold, S., Kersten, M.L.: Database architecture optimized for the new bottleneck: memory access. In: 25th International Conference on Very Large Data Bases (1999)Google Scholar
  3. 3.
    Dreseler, M., et al.: Hardware-accelerated memory operations on large-scale NUMA systems. In: Eighth International Workshop on Accelerating Analytics and Data Management Systems Using Modern Processor and Storage Architectures (ADMS) (2017)Google Scholar
  4. 4.
    Foong, A.P., et al.: TCP performance re-visited. In: IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) (2003)Google Scholar
  5. 5.
    Neumann, T., Mühlbauer, T., Kemper, A.: Fast serializable multi-version concurrency control for main-memory database systems. In: ACM SIGMOD International Conference on Management of Data (2015)Google Scholar
  6. 6.
    Oracle. Oracle’s SPARC T7 and SPARC M7 Server Architecture. Technical report. Accessed 12 June 2017
  7. 7.
    Psaroudakis, I., et al.: Adaptive NUMA-aware data placement and task scheduling for analytical workloads in main-memory column-stores. In: Proceedings of the VLDB (2016)Google Scholar
  8. 8.
    Thorson, G., Woodacre, M.: SGI UV2: a fused computation and data analysis machine. In: International Conference on High Performance Computing, Networking, Storage and Analysis (2012)Google Scholar
  9. 9.
    Ungethüm, A., et al.: Overview on hardware optimizations for database engines. In: Datenbanksysteme für Business, Technologie und Web (BTW) (2017)Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Markus Dreseler
    • 1
    Email author
  • Timo Gasda
    • 1
  • Jan Kossmann
    • 1
  • Matthias Uflacker
    • 1
  • Hasso Plattner
    • 1
  1. 1.Hasso Plattner InstitutePotsdamGermany

Personalised recommendations