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:20260422T143138Z
LOCATION:B309
DTSTART;TZID=America/New_York:20241121T113000
DTEND;TZID=America/New_York:20241121T120000
UID:submissions.supercomputing.org_SC24_sess390_pap175@linklings.com
SUMMARY:A Probabilistic Approach to Selecting Build Configurations in Pack
 age Managers
DESCRIPTION:Daniel Nichols (University of Maryland), Harshitha Menon and T
 odd Gamblin (Lawrence Livermore National Laboratory (LLNL)), and Abhinav B
 hatele (University of Maryland)\n\nModern scientific software in high perf
 ormance computing is often complex, and many parallel applications and lib
 raries depend on several other software or libraries.  Developers and user
 s of such complex software often use package managers for building them.  
 Package managers depend on humans to codify package constraints, and the d
 ependency graph of a software package can often become large. In this pape
 r, we propose a methodology that uses historical build results to assist a
  package manager in selecting the best versions of package dependencies wi
 th an aim to improve the likelihood of a successful build.  We train a mac
 hine learning (ML) model to predict the probability of build outcomes of d
 ifferent configurations of packages in the Spack package manager.  When ev
 aluated on common scientific software stacks, this ML model-based approach
  is able to achieve a 13% higher success rate in building packages than th
 e default version selection mechanism in Spack.\n\nTag: Accelerators, Arti
 ficial Intelligence/Machine Learning, Codesign, State of the Practice, Sys
 tem Administration\n\nRegistration Category: Tech Program Reg Pass\n\nSess
 ion Chair: George Michelogiannakis (Lawrence Berkeley National Laboratory 
 (LBNL), Stanford University)\n\n
END:VEVENT
END:VCALENDAR