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:20260422T000604Z
LOCATION:DEF Concourse
DTSTART;TZID=America/Denver:20231115T100000
DTEND;TZID=America/Denver:20231115T170000
UID:submissions.supercomputing.org_SC23_sess303_rpost197@linklings.com
SUMMARY:Transfer Learning Workflow for High-Quality I/O Bandwidth Predicti
 on with Limited Data
DESCRIPTION:Dmytro Povaliaiev (RWTH Aachen University); Radita Liem (RWTH 
 Aachen University, IT Center); Julian Kunkel (University of Göttingen, GWD
 G, Germany); Jay Lofstead (Sandia National Laboratories); and Philip Carns
  (Argonne National Laboratory (ANL))\n\nThe I/O performance prediction is 
 challenging due to multiple intertwined variables inside a cluster. This s
 ituation makes I/O performance prediction a strong candidate for using mac
 hine learning because of the complex variables involved. However, making a
  high-quality prediction requires a large amount of equivalent-quality dat
 a, and collecting it is a big challenge for most data centers.\n\nIn this 
 project, we explore transfer learning to predict the I/O performance by ut
 ilizing the publicly available I/O performance data in Darshan logs from t
 he NCSA's Blue Waters supercomputer. We devise a workflow to train a neura
 l network model as a base to predict the POSIX I/O bandwidth of other clus
 ters (CLAIX18 and Theta). With less than 1% of the data needed to build th
 e base model, our experiment shows that our transfer learning workflow can
  predict the I/O bandwidth of another system with a mean absolute error be
 tter or equivalent to the state-of-the-art.\n\nTag: Artificial Intelligenc
 e/Machine Learning, Architecture and Networks, Heterogeneous Computing, I/
 O and File Systems, Performance Measurement, Modeling, and Tools, Post-Moo
 re Computing, Programming Frameworks and System Software, Quantum Computin
 g\n\nRegistration Category: Tech Program Reg Pass, Exhibits Reg Pass\n\n
END:VEVENT
END:VCALENDAR