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:20181221T160728Z
LOCATION:D161
DTSTART;TZID=America/Chicago:20181112T142000
DTEND;TZID=America/Chicago:20181112T144000
UID:submissions.supercomputing.org_SC18_sess158_ws_lasalss106@linklings.co
 m
SUMMARY:Shift-Collapse Acceleration of Generalized Polarizable Reactive Mo
 lecular Dynamics for Machine Learning-Assisted Computational Synthesis of 
 Layered Materials
DESCRIPTION:Workshop\nAlgorithms, Heterogeneous Systems, Resiliency, Works
 hop Reg Pass\n\nShift-Collapse Acceleration of Generalized Polarizable Rea
 ctive Molecular Dynamics for Machine Learning-Assisted Computational Synth
 esis of Layered Materials\n\nLiu, Tiwari, Sheng, Krishnamoorthy, Hong...\n
 \nReactive molecular dynamics is a powerful simulation method for describi
 ng chemical reactions. Here, we introduce a new generalized polarizable re
 active force-field (ReaxPQ+) model to significantly improve the accuracy b
 y accommodating the reorganization of surrounding media. The increased com
 putation is accelerated by (1) extended Lagrangian approach to eliminate t
 he speed-limiting charge iteration, (2) shiftcollapse computation of many-
 body renormalized n-tuples, which provably minimizes data transfer, (3) mu
 ltithreading with roundrobin data privatization, and (4) data reordering t
 o reduce computation and allow vectorization. The new code achieves (1) we
 ak-scaling parallel efficiency of 0.989 for 131,072 cores, and (2) eight-f
 old reduction of time-to-solution (T2S) compared with the original code, o
 n an Intel Knights Landing-based computer. The reduced T2S has for the fir
 st time allowed purely computational synthesis of atomically-thin transiti
 on metal dichalcogenide layers assisted by machine learning to discover a 
 novel synthetic pathway.
URL:https://sc18.supercomputing.org/presentation/?id=ws_lasalss106&sess=se
 ss158
END:VEVENT
END:VCALENDAR