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:20260202T201803Z
LOCATION:261
DTSTART;TZID=America/Chicago:20251116T104500
DTEND;TZID=America/Chicago:20251116T110000
UID:submissions.supercomputing.org_SC25_sess220_ws_eduhpcp112@linklings.co
 m
SUMMARY:The Cost of Teaching Operational ML
DESCRIPTION:Fraida Fund (New York University); Kate Keahey (Argonne Nation
 al Laboratory (ANL)); Cody Hammock (Texas Advanced Computing Center (TACC)
 ); and Marc Richardson, Mark Powers, and Michael Sherman (University of Ch
 icago)\n\nOperational machine learning (ML) requires skills beyond model d
 evelopment, including infrastructure provisioning, large-scale training ac
 ross clusters, model deployment with consideration of operational performa
 nce, monitoring, and automation - capabilities grounded in high-performanc
 e computing and distributed systems. This paper presents the design and in
 frastructure requirements of a graduate-level course on ML Systems Enginee
 ring and Operations, aimed at equipping students with these skills. Using 
 186,692 total compute instance hours on the Chameleon Cloud testbed, stude
 nts built end-to-end ML pipelines incorporating distributed training, repr
 oducible experiment tracking, automated re-training and re-deployment, and
  continuous monitoring. We analyze compute usage across assignments, compa
 re expected versus actual resource consumption, and estimate that replicat
 ing the course on commercial cloud platforms would cost approximately $250
  per student (almost $50,000 for our course with enrollment of 191 student
 s). \nAll course materials are publicly available for reuse.\n\nRecording:
  Livestreamed, Recorded\n\nRegistration Category: Technical Program Reg Pa
 ss, Workshop Reg Pass\n\nSession Chairs: Erik Saule (University of North C
 arolina at Charlotte), Sushil K. Prasad (University of Texas at San Antoni
 o), David P. Bunde (Knox College), and Suzanne J. Matthews (United States 
 Military Academy)\n\n
END:VEVENT
END:VCALENDAR