BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160555Z
LOCATION:Track 8
DTSTART;TZID=America/New_York:20201112T133000
DTEND;TZID=America/New_York:20201112T135500
UID:submissions.supercomputing.org_SC20_sess214_ws_lasalss109@linklings.co
m
SUMMARY:A Survey of Singular Value Decomposition Methods for Distributed
Tall/Skinny Data
DESCRIPTION:Workshop\n\nA Survey of Singular Value Decomposition Methods f
or Distributed Tall/Skinny Data\n\nSchmidt\n\nThe Singular Value Decompos
ition (SVD) is one of the most important matrix \nfactorizations, enjoying
a wide variety of applications across numerous \napplication domains. In
statistics and data analysis, the common applications of \nSVD such as Pri
ncipal Components Analysis (PCA) and linear regression. Usually \nthese ap
plications arise on data that has far more rows than columns, so-called\n"
tall/skinny" matrices. In the big data analytics context, this may take th
e \nform of hundreds of millions to billions of rows with only a few hundr
ed \ncolumns. There is a need, therefore, for fast, accurate, and scalable
\ntall/skinny SVD implementations which can fully utilize modern computin
g \nresources. To that end, we present a survey of three different algorit
hms for \ncomputing the SVD for these kinds of tall/skinny data layouts us
ing MPI for \ncommunication. We contextualize these with common big data a
nalytics \ntechniques, principally PCA. Finally, we present both CPU and G
PU timing \nresults from the Summit supercomputer, and discuss possible al
ternative \napproaches.\n\nTag: Algorithms, Extreme Scale Computing, Perfo
rmance/Productivity Measurement and Evaluation, Scalable Computing, Scient
ific Computing\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR