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:20260202T201247Z
LOCATION:Second Floor Atrium
DTSTART;TZID=America/Chicago:20251120T080000
DTEND;TZID=America/Chicago:20251120T170000
UID:submissions.supercomputing.org_SC25_sess533_post151@linklings.com
SUMMARY:Mojo: Python-Like MLIR-Based GPU Portable Science Kernels
DESCRIPTION:Tatiana Melnichenko (University of Tennessee, Knoxville; Oak R
 idge National Laboratory (ORNL))\n\nThis work investigates Mojo, a new MLI
 R-based language that combines Python-like syntax with portable, low-level
  GPU programming capabilities. We compare the performance of the Mojo port
 able GPU kernels against vendor-specific C++ NVIDIA CUDA and AMD HIP imple
 mentations on four representative scientific workloads: (1) BabelStream (m
 emory-bound); (2) seven-point stencil (memory-bound); (3) miniBUDE (comput
 e-bound); and (4) Hartree-Fock (compute-bound with atomic operations), eva
 luated on NVIDIA H100 and AMD MI300A GPUs. Results show that Mojo can matc
 h CUDA and HIP performance for memory-bound kernels, though gaps remain fo
 r atomic operations and certain compute-bound cases. This poster will pres
 ent a general overview of the language, our benchmarking methodology, comp
 arative results, the use of vendor profiling tools, and observations on Mo
 jo’s potential to close the gap between high performance and developer pro
 ductivity in scientific GPU programming. Our contribution is the first sys
 tematic evaluation of Mojo for HPC workloads, highlighting both its promis
 e and current limitations.\n\nTag: Research & ACM SRC Posters\n\nRegistrat
 ion Category: Technical Program Reg Pass\n\nSession Chairs: Kento Sato (RI
 KEN Center for Computational Science (R-CCS)); Chris Schlipalius (Pawsey S
 upercomputing Research Centre; Commonwealth Scientific and Industrial Rese
 arch Organisation (CSIRO), Australia); and Anja Gerbes (Georg-August-Unive
 rsität Göttingen)\n\n
END:VEVENT
END:VCALENDAR