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:20211207T055343Z
LOCATION:Online
DTSTART;TZID=America/Chicago:20211114T153000
DTEND;TZID=America/Chicago:20211114T160000
UID:submissions.supercomputing.org_SC21_sess432_ws_waccpd106@linklings.com
SUMMARY:Achieving near native runtime performance and cross-platform perfo
 rmance portability for random number generation through SYCL interoperabil
 ity
DESCRIPTION:Workshop\n\nAchieving near native runtime performance and cros
 s-platform performance portability for random number generation through SY
 CL interoperability\n\nPascuzzi, Goli\n\nHigh-performance computing (HPC) 
 is a major driver accelerating scientific research and discovery, from qua
 ntum simulations to medical therapeutics. The growing number of new HPC sy
 stems coming online are being furnished with various hardware components, 
 engineered by competing industry entities, each having their own architect
 ures and platforms to be supported. While the increasing availability of t
 hese resources is in many cases pivotal to successful science, even the la
 rgest collaborations lack the computational expertise required for maximal
  exploitation of current hardware capabilities. The need to maintain multi
 ple platform-specific codebases further complicates matters, potentially a
 dding a constraint on the number of machines that can be utilized. Fortuna
 tely, numerous programming models are under development that aim to facili
 tate software solutions for heterogeneous computing. In particular is SYCL
 , an open standard, C++-based single-source programming paradigm. Among SY
 CL's features is interoperability, a mechanism through which applications 
 and third-party libraries coordinate  sharing data and execute collaborati
 vely. In this paper, we leverage the SYCL programming model to demonstrate
  cross-platform performance portability across heterogeneous resources. We
  detail our NVIDIA and AMD random number generator extensions to the oneMK
 L open-source interfaces library. Performance portability is measured rela
 tive to platform-specific baseline applications executed on four major har
 dware platforms using two different compilers supporting SYCL. The utility
  of our extensions are exemplified in a real-world setting via a high-ener
 gy physics simulation application. We show the performance of implementati
 ons that capitalize on SYCL interoperability are at par with native implem
 entations, attesting to the cross-platform performance portability of a SY
 CL-based approach to scientific codes.\n\nTag: Online Only, Accelerator-ba
 sed Architectures, Parallel Programming Languages and Models, Performance,
  State of the Practice\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR