BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20211207T055404Z
LOCATION:Online
DTSTART;TZID=America/Chicago:20211114T153000
DTEND;TZID=America/Chicago:20211114T160000
UID:submissions.supercomputing.org_SC21_sess430_ws_exampi107@linklings.com
SUMMARY:Accelerating Multi-Process Communication for Parallel 3-D FFT
DESCRIPTION:Workshop\n\nAccelerating Multi-Process Communication for Paral
 lel 3-D FFT\n\nAyala, Tomov, Stoyanov, Haidar, Dongarra\n\nToday's largest
  and most powerful supercomputers in the world are built on heterogeneous 
 platforms; and using the combined power of multi-core CPUs and GPUs, has h
 ad a great impact accelerating large-scale applications. However, on these
  architectures, parallel algorithms, such as the Fast Fourier Transform (F
 FT), encounter that inter-processor communication become a bottleneck and 
 limits their scalability. In this paper, we present techniques for speedin
 g up multi-process communication cost during the computation of FFTs, cons
 idering hybrid network connections as those expected on upcoming exascale 
 machines. Among our techniques, we present algorithmic tuning, making use 
 of phase diagrams; parametric tuning, using different FFT settings; and MP
 I distribution tuning based on FFT size and computational resources availa
 ble. We present several experiments obtained on Summit supercomputer at Oa
 k Ridge National Laboratory, using up to 40,960 IBM Power9 cores and 6,144
  NVIDIA V-100 GPUs.\n\nTag: Online Only, Extreme Scale Computing, Parallel
  Programming Languages and Models, Performance\n\nRegistration Category: W
 orkshop Reg Pass
END:VEVENT
END:VCALENDAR