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:20200129T163226Z
LOCATION:301-302-303
DTSTART;TZID=America/Denver:20191120T133000
DTEND;TZID=America/Denver:20191120T140000
UID:submissions.supercomputing.org_SC19_sess142_pap149@linklings.com
SUMMARY:Conflict-Free Symmetric Sparse Matrix-Vector Multiplication on Mul
ticore Architectures
DESCRIPTION:Paper\n\nConflict-Free Symmetric Sparse Matrix-Vector Multipli
cation on Multicore Architectures\n\nElafrou, Goumas, Koziris\n\nExploitin
g the numeric symmetry in sparse matrices to reduce their memory footprint
is very tempting for optimizing the memory-bound Sparse Matrix-Vector Mul
tiplication (SpMV) kernel. Despite being very beneficial for serial comput
ation, storing the upper or lower triangular part of the matrix introduces
race conditions in the updates to the output vector in a parallel executi
on. Previous work has suggested using local, per-thread vectors to circumv
ent this problem, introducing a work-inefficient reduction step that limit
s the scalability of SpMV. In this paper, we address this issue with Confl
ict-Free Symmetric (CFS) SpMV, an optimization strategy that organizes the
parallel computation into phases of conflict-free execution. We identify
such phases through graph coloring and propose heuristics to improve the c
oloring quality for SpMV in terms of load balancing and locality to the in
put and output vectors. We evaluate our approach on two multicore shared-m
emory systems and demonstrate improved performance over the state-of-the-a
rt.\n\nTag: Tech Program Reg Pass, Algorithms, Linear Algebra, Performance
, Sparse Computation\n\nRegistration Category: Tech Program Reg Pass, Algo
rithms, Linear Algebra, Performance, Sparse Computation
URL:https://sc19.supercomputing.org/presentation/?id=pap149&sess=sess142
END:VEVENT
END:VCALENDAR