BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.riscv-europe.org//eu-summit-2026//speaker//M3YPYK
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-eu-summit-2026-HNQPKX@cfp.riscv-europe.org
DTSTART;TZID=CET:20260611T123000
DTEND;TZID=CET:20260611T124500
DESCRIPTION:We implemented an N-body astrophysical simulation code and offl
 oaded its most computationally intensive kernel to Tenstorrent RISC-V–ba
 sed accelerators using the TT-Metalium programming interface. Performance 
 was assessed on the Wormhole n300 card in terms of execution time and ener
 gy consumption\, and compared with both an optimized CPU implementation an
 d a CUDA version. The TT-Metalium implementation achieves a speedup of 2×
  over the CPU baseline\, although its performance still slightly lags behi
 nd the CUDA implementation. Finally\, we investigated strategies for scali
 ng the application across multiple Tenstorrent accelerators\, evaluating c
 onfigurations with up to four devices.
DTSTAMP:20260522T162430Z
LOCATION:Plenary
SUMMARY:Evaluating Tenstorrent RISC-V Accelerators for High Performance Sci
 entific Computing - Elisabetta Boella
URL:https://cfp.riscv-europe.org/eu-summit-2026/talk/HNQPKX/
END:VEVENT
END:VCALENDAR