TU Berlin

Fachgebiet Datenbanksysteme und InformationsmanagementPublikationen

Logo FG DIMA-new  65px


zur Navigation


Triton Join: Efficiently Scaling the Operator State on GPUs with Fast Interconnects
Zitatschlüssel LutzBZRM22
Autor Clemens Lutz, Sebastian Breß, Steffen Zeuch, Tilmann Rabl, Volker Markl
Jahr 2022
Journal SIGMOD
Notiz to appear
Zusammenfassung Database management systems are facing growing data volumes. Previous research suggests that GPUs are well-equipped to quickly process joins and similar stateful operators, as GPUs feature high-bandwidth on-board memory. However, GPUs cannot scale joins and similar stateful operators to large data volumes due to two limiting factors: (1) large state does not fit into the on-board memory, and (2) spilling state to main memory is constrained by the interconnect bandwidth. Thus, CPUs are often still the better choice for scalable data processing. In this paper, we propose a new join algorithm that scales to large data volumes by taking advantage of fast interconnects. Fast interconnects such as NVLink 2.0 are a new technology that connect the GPU to main memory at a high bandwidth, and thus enable us to design our join to efficiently spill its operator state. Our evaluation shows that our Triton join outperforms a no-partitioning hash join by more than 100× on the same GPU, and a radix-partitioned join on the CPU by up to 2.5×. As a result, GPU-enabled DBMSs are able to scale beyond the GPU memory capacity.
Link zur Publikation Download Bibtex Eintrag



Schnellnavigation zur Seite über Nummerneingabe