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:20260422T000713Z
LOCATION:702
DTSTART;TZID=America/Denver:20231112T153000
DTEND;TZID=America/Denver:20231112T160000
UID:submissions.supercomputing.org_SC23_sess436_ws_ia106@linklings.com
SUMMARY:Towards a Massive-Scale Distributed Neighborhood Graph Constructio
 n
DESCRIPTION:Keita Iwabuchi, Trevor Steil, Benjamin Priest, Roger Pearce, a
 nd Geoffrey Sanders (Lawrence Livermore National Laboratory (LLNL))\n\nGra
 ph-based approximate nearest neighbor algorithms have shown high performan
 ce and quality.  However, such approaches require a large amount of memory
  and still take a long time to construct high-quality nearest neighbor gra
 phs (NNGs).  Using distributed memory systems is important when data is la
 rge or a shorter indexing time is desired.\n\nWe develop a distributed mem
 ory version of NN-Descent, a widely known graph-based ANN algorithm, close
 ly following algorithmic advances by PyNN-Descent authors.  Our distribute
 d NN-Descent (DNND) is built on top of MPI and leverages two existing high
 -performance computing libraries: YGM (an asynchronous communication libra
 ry) and Metall (a persistent memory allocator).\n\nWe evaluate the perform
 ance of DNND on an HPC system using billion-scale datasets, demonstrating 
 that our approach shows high performance and strong scaling and has great 
 potential for developing massive-scale NNG frameworks.\n\nTag: Algorithms,
  Applications, Architecture and Networks\n\nRegistration Category: Worksho
 p Reg Pass\n\nSession Chairs: Sanjukta Bhowmick (University of North Texas
 ), Vito Giovanni Castellana (Pacific Northwest National Laboratory (PNNL))
 , John Feo (Pacific Northwest National Laboratory (PNNL)), Marco Minutoli 
 (Advanced Micro Devices), and Antonino Tumeo (Pacific Northwest National L
 aboratory (PNNL))\n\n
END:VEVENT
END:VCALENDAR