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:20181221T160729Z
LOCATION:C140/142
DTSTART;TZID=America/Chicago:20181113T143000
DTEND;TZID=America/Chicago:20181113T150000
UID:submissions.supercomputing.org_SC18_sess206_pap120@linklings.com
SUMMARY:Redesigning LAMMPS for Petascale and Hundred-Billion-Atom Simulati
 on on Sunway TaihuLight
DESCRIPTION:Paper\nAlgorithms, Applications, Computational Biology, Scient
 ific Computing, Tech Program Reg Pass\n\nRedesigning LAMMPS for Petascale 
 and Hundred-Billion-Atom Simulation on Sunway TaihuLight\n\nDuan, Gao, Zha
 ng, Zhang, Liu...\n\nLarge-scale molecular dynamics (MD) simulations on su
 percomputers play an increasingly important role in many research areas. I
 n this paper, we present our efforts on redesigning the widely used LAMMPS
  MD simulator for Sunway TaihuLight supercomputer and its ShenWei many-cor
 e architecture (SW26010). The memory constraints of SW26010 bring a number
  of new challenges for achieving efficient MD implementation on it. In ord
 er to overcome these constraints, we employ four levels of optimization: (
 1) a hybrid memory update strategy; (2) a software cache strategy; (3) cus
 tomized transcendental math functions; and (4) a full pipeline acceleratio
 n. Furthermore, we redesign the code to enable all possible vectorization.
  Experiments show that our redesigned software on a single SW26010 process
 or can outperform over 100 E5-2650 cores for running the latest stable rel
 ease (11Aug17) of LAMMPS. We also achieve a performance of over 2.43 PFlop
 s for a Tersoff simulation when using 16,384 nodes on Sunway TaihuLight.
URL:https://sc18.supercomputing.org/presentation/?id=pap120&sess=sess206
END:VEVENT
END:VCALENDAR

