BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000712Z
LOCATION:605
DTSTART;TZID=America/Denver:20231113T104800
DTEND;TZID=America/Denver:20231113T105700
UID:submissions.supercomputing.org_SC23_sess441_ws_p3hpc111@linklings.com
SUMMARY:Porting Batched Iterative Solvers onto Intel GPUs with SYCL
DESCRIPTION:Phuong Nguyen (University of Tennessee, Innovative Computing L
 aboratory (ICL)); Pratik Nayak (Karlsruhe Institute of Technology (KIT)); 
 and Hartwig Anzt (University of Tennessee, Innovative Computing Laboratory
  (ICL))\n\nBatched linear solvers play a vital role in computational scien
 ces, especially in the fields of plasma physics and combustion simulations
 . With the imminent deployment of the Aurora Supercomputer and other upcom
 ing systems equipped with Intel GPUs, there is a compelling demand to expa
 nd the capabilities of these solvers for Intel GPU architectures.\n\nWe pr
 esent our efforts in porting and optimizing the batched iterative solvers 
 on Intel GPUs using the SYCL programming model. These new solvers achieve 
 impressive performance on the Intel GPU Max 1550s (Ponte Vecchio GPUs) whi
 ch surpass our previous CUDA implementation on NVIDIA H100 GPUs by an aver
 age of 2.4x for the PeleLM application inputs. The batched solvers are rea
 dy for production use in real-world scientific applications through the Gi
 nkgo library, complementing the performance portability of the batched fun
 ctionality of Ginkgo.\n\nTag: Performance Measurement, Modeling, and Tools
 , Performance Optimization\n\nRegistration Category: Workshop Reg Pass\n\n
 Session Chairs: Judith C. Hill (Lawrence Livermore National Laboratory (LL
 NL)), CJ Newburn (NVIDIA Corporation), Scott J. Parker (Argonne National L
 aboratory (ANL)), and John Pennycook (Intel Corporation)\n\n
END:VEVENT
END:VCALENDAR