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:20250626T234540Z
LOCATION:B314
DTSTART;TZID=America/New_York:20241118T111500
DTEND;TZID=America/New_York:20241118T120000
UID:submissions.supercomputing.org_SC24_sess758_misc380@linklings.com
SUMMARY:Architectural Challenges in Modern AI Workloads
DESCRIPTION:Pavan Balaji (Meta)\n\nLike many other hyperscalar companies, 
 Meta heavily relies on AI in its internal systems for numerous things incl
 uding recommending personalized content to the user; understanding text, s
 peech, and visual content to recognize things such as hate speech; and gen
 erating text, image and video content for users. While these AI workloadss
 hare some similarities with traditional scientific computing applications,
  they are a lot less structured---their computational requirements, memory
  access patterns, data I/O approaches, and network communication, tend to 
 be highly irregular raising tremendous challenges on modern supercomputing
  system architectures. In this talk, I'll discuss some of these architectu
 ral challenges and allude to some potential hardware and software solution
 s that Meta as well as the broader community is looking at to address thes
 e challenges.\n\nTag: Artificial Intelligence/Machine Learning, Codesign\n
 \nRegistration Category: Workshop Reg Pass\n\nSession Chairs: John Feo (Pa
 cific Northwest National Laboratory (PNNL)), Jiyuan Zhang (Meta), and Amel
 ie Chi Zhou (Hong Kong Baptist University)\n\n
END:VEVENT
END:VCALENDAR