BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000711Z
LOCATION:403-404
DTSTART;TZID=America/Denver:20231114T133000
DTEND;TZID=America/Denver:20231114T140000
UID:submissions.supercomputing.org_SC23_sess165_pap319@linklings.com
SUMMARY:The Graph Database Interface: Scaling Online Transactional and Ana
 lytical Graph Workloads to Hundreds of Thousands of Cores
DESCRIPTION:Maciej Besta, Robert Gerstenberger, and Marc Fischer (ETH Zuri
 ch - Swiss Federal Institute of Technology); Michał Podstawski (TCL Eagle 
 Lab, Warsaw University of Technology); Nils Blach, Berke Egeli, and Georgy
  Mitenkov (ETH Zurich - Swiss Federal Institute of Technology); Wojciech C
 hlapek (ICM UW); Marek Michalewicz (Sano Centre for Computational Medicine
 , Krakow, Poland); Hubert Niewiadomski (Cledar); Jürgen Müller (BASF SE); 
 and Torsten Hoefler (ETH Zurich - Swiss Federal Institute of Technology)\n
 \nGraph databases (GDBs) are crucial in academic and industry applications
 . The key challenges in developing GDBs are achieving high performance, sc
 alability, programmability, and portability. To tackle these challenges, w
 e harness established practices from the HPC landscape to build a system t
 hat outperforms all past GDBs presented in the literature by orders of mag
 nitude, for both OLTP and OLAP workloads. For this, we first identify and 
 crystallize performance-critical building blocks in the GDB design, and ab
 stract them into a portable and programmable API specification, called the
  Graph Database Interface (GDI), inspired by the best practices of MPI. We
  then use GDI to design a GDB for distributed-memory RDMA architectures. O
 ur implementation harnesses one-sided RDMA communication and collective op
 erations, and it offers architecture-independent theoretical performance g
 uarantees. The resulting design achieves extreme scales of more than a hun
 dred thousand cores. Our work will facilitate the development of next-gene
 ration extreme-scale graph databases.\n\nTag: Cloud Computing, Data Analys
 is, Visualization, and Storage, Graph Algorithms and Frameworks\n\nRegistr
 ation Category: Tech Program Reg Pass\n\nAward Finalist: Best Paper Finali
 st\n\nReproducibility Badges: Artifact Available, Artifact Functional, Res
 ults Reproduced\n\nSession Chair: Kasimir Gabert (Sandia National Laborato
 ries)\n\n
END:VEVENT
END:VCALENDAR