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:20260422T000713Z
LOCATION:505
DTSTART;TZID=America/Denver:20231113T143000
DTEND;TZID=America/Denver:20231113T143300
UID:submissions.supercomputing.org_SC23_sess442_ws_whpc110@linklings.com
SUMMARY:Queue Wait Time Prediction in Supercomputers
DESCRIPTION:Nwamaka Okafor and Bethany Lusch (Argonne National Laboratory 
 (ANL))\n\nHigh Performance Computing systems play critical role in advanci
 ng scientific research. They use schedulers for allocating resources to qu
 eued jobs. Waiting time can vary, even among jobs of similar characteristi
 cs, making it difficult to accurately estimate the exact time a job will w
 ait in the queue. Knowing how long a job will wait is beneficial for adequ
 ate planning and to avoid frustrations that may arise when a user's expect
 ation of waiting time is not met. Efficient job wait time estimation is al
 so crucial for optimizing resource allocation.\n\nThis work investigates t
 he impact of job characteristics and user behaviors on job wait time on  l
 eadership-class HPC systems. The paper evaluates the performance of differ
 ent supervised learning algorithms for job wait time estimation.  While th
 is study focuses on the workload and hardware characteristics of Theta Cra
 yXC40, the processes and tools developed in this study can be applied to a
 ny other leadership-class machine.\n\nTag: State of the Practice\n\nRegist
 ration Category: Workshop Reg Pass\n\nSession Chairs: Elsa J. Gonsiorowski
  (Lawrence Livermore National Laboratory (LLNL)) and Mozhgan Kabiri chimeh
  (NVIDIA Corporation)\n\n
END:VEVENT
END:VCALENDAR