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:20260422T143139Z
LOCATION:B206
DTSTART;TZID=America/New_York:20241119T113000
DTEND;TZID=America/New_York:20241119T120000
UID:submissions.supercomputing.org_SC24_sess678_exforum131@linklings.com
SUMMARY:New Optical-Based Scale-Up Fabrics to Meet the Performance and TCO
  Requirements of Next-Generation AI/HPC Architectures
DESCRIPTION:Vladimir Stojanovic (Ayar Labs) and Atsushi Miki (Fujitsu Ltd)
 \n\nAs large language models rapidly grow, compute demands have surged, ge
 nerating staggering amounts of data and requiring high-speed data movement
  between AI accelerators, memory and storage. Conventional interconnects a
 nd optics fail to scale to the cost and throughput that AI/HPC systems wil
 l require, causing major bottlenecks, poor hardware use, higher power cons
 umption, and increased costs. New connectivity solutions are needed for ef
 ficient, high-performance AI/HPC architectures.\n\nIn this session, Fujits
 u will highlight current AI/HPC architecture challenges and future needs. 
 Ayar Labs will present a chiplet-based optical I/O solution offering signi
 ficantly better bandwidth, power efficiency, and latency compared to tradi
 tional methods. They will also share new analysis on how optical I/O impro
 ves system performance and total cost of ownership (TCO) for AI inference 
 and training. Attendees will learn how these new optical fabrics meet the 
 performance and TCO requirements of next-gen AI/HPC systems.\n\nTag: Netwo
 rk\n\nRegistration Category: Tech Program Reg Pass, Exhibits Reg Pass\n\nS
 ession Chair: Ben Bergen (Los Alamos National Laboratory (LANL), Applied C
 omputer Science Group)\n\n
END:VEVENT
END:VCALENDAR
