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:C146
DTSTART;TZID=America/Chicago:20181113T143000
DTEND;TZID=America/Chicago:20181113T150000
UID:submissions.supercomputing.org_SC18_sess178_pap141@linklings.com
SUMMARY:Distributed-Memory Hierarchical Compression of Dense SPD Matrices
DESCRIPTION:Paper\nAlgorithms, Architectures, Data Analytics, Deep Learnin
 g, Networks, Scientific Computing, Visualization, Tech Program Reg Pass, B
 SP Finalist\n\nDistributed-Memory Hierarchical Compression of Dense SPD Ma
 trices\n\nYu, Reiz, Biros\n\nWe present a distributed-memory algorithm for
  the hierarchical compression of SPD matrices. Our method is based on GOFM
 M, an algorithm that appeared in doi:10.1145/3126908.3126921.\n\nFor many 
 SPD matrices, GOFMM enables compression and approximate matrix-vector mult
 iplication in NlogN time---as opposed to quadratic work required for a den
 se matrix. But GOFMM supports only shared memory parallelism. In this pape
 r, we use the message passing interface, extending the ideas of GOFMM to t
 he distributed memory setting. We also introduce an asynchronous algorithm
  for faster multiplication. We present different usage scenarios of SPD ma
 trices that are related to graphs, neural-networks, and covariance operato
 rs. We also compare with STRUMPACK, which, to our knowledge, is the only o
 ther parallel software that can compress arbitrary SPD matrices. In our la
 rgest run, we were able to compress a 67M-by-67M matrix within three minut
 es and perform a multiplication with 512 vectors within 5 seconds on 6,144
  Intel Skylake cores.
URL:https://sc18.supercomputing.org/presentation/?id=pap141&sess=sess178
END:VEVENT
END:VCALENDAR

