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:20260924T172458Z
LOCATION:B312
DTSTART;TZID=America/New_York:20241118T110000
DTEND;TZID=America/New_York:20241118T113000
UID:submissions.supercomputing.org_SC24_sess811_ws_ciss102@linklings.com
SUMMARY:Benchmarking Ethernet Interconnect for HPC/AI workloads
DESCRIPTION:Lorenzo Pichetti (University of Trento, Italy); Daniele De Sen
 si (Sapienza University of Rome); Karthee Sivalingam (Huawei Technologies 
 Ltd, KTH Royal Institute of Technology); Stepan Nassyr (ParTec AG, Germany
 ; Forschungszentrum Jülich); Dirk Pleiter (KTH Royal Institute of Technolo
 gy); Aldo Artigiani (Huawei Technologies Ltd); Flavio Vella (University of
  Trento, Italy); and Daniele Cesarini and Matteo Turisini (CINECA)\n\nInte
 rconnects have always played a cornerstone role\nin HPC. Since the incepti
 on of the Top500 ranking, interconnect\nstatistics have been predominantly
  dominated by two compet-\ning technologies: InfiniBand and Ethernet. Howe
 ver, even if\nEthernet increased its popularity due to versatility and cos
 t-\neffectiveness, InfiniBand used to provide higher bandwidth and\ncontin
 ues to feature lower latency. Industry seeks for a further\nevolution of t
 he Ethernet standards to enable fast and low-\nlatency interconnect for em
 erging AI workloads by offering\ncompetitive, open-standard solutions. Thi
 s paper analyzes the\nearly results obtained from two systems relying on a
 n HPC\nEthernet interconnect, one relying on 100G and the other on\n200G E
 thernet. Preliminary findings indicate that the Ethernet-\nbased networks 
 exhibit competitive performance, closely aligning\nwith InfiniBand, especi
 ally for large message exchanges.\n\nTag: I/O, Storage, Archive\n\nRegistr
 ation Category: Workshop Reg Pass\n\nSession Chairs: Glenn Brook (Cornelis
  Networks, University of Tennessee); Clayton Hughes (Sandia National Labor
 atories); Nalini Kumar (Intel Corporation); Hatem Ltaief (King Abdullah Un
 iversity of Science and Technology (KAUST)); David Martin (Lawrence Berkel
 ey National Laboratory (LBNL), Energy Sciences Network (ESnet)); and Amit 
 Ruhela (Texas Advanced Computing Center (TACC), University of Texas)\n\n
END:VEVENT
END:VCALENDAR