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:20230124T171520Z
LOCATION:C140-142
DTSTART;TZID=America/Chicago:20221115T153000
DTEND;TZID=America/Chicago:20221115T160000
UID:submissions.supercomputing.org_SC22_sess172_pap430@linklings.com
SUMMARY:CA3DMM: A New Algorithm Based on a Unified View of Parallel Matrix
Multiplication
DESCRIPTION:Paper\n\nCA3DMM: A New Algorithm Based on a Unified View of Pa
rallel Matrix Multiplication\n\nHuang, Chow\n\nThis paper presents the Com
munication-Avoiding 3D Matrix Multiplication (CA3DMM) algorithm, a simple
and novel algorithm that has optimal or near-optimal communication cost. C
A3DMM is based on a unified view of parallel matrix multiplication. Such a
view generalizes 1D, 2D, and 3D matrix multiplication algorithms to reduc
e the data exchange volume for different shapes of input matrices. CA3DMM
further minimizes the actual communication costs by carefully organizing i
ts communication patterns. CA3DMM is much simpler than some other generali
zed 3D algorithms, and CA3DMM does not require low-level optimization. Num
erical experiments show that CA3DMM has good parallel scalability and has
similar or better performance when compared to state-of-the-art PGEMM impl
ementations for a wide range of matrix dimensions and number of processes.
\n\nSession Format: Recorded\n\nTag: Numerical Algorithms, Scientific Comp
uting\n\nRegistration Category: Tech Program Reg Pass\n\nReproducibility B
adges: Artifact Available, Artifact Functional, Results Reproduced
END:VEVENT
END:VCALENDAR