BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20241120T082410Z
LOCATION:HG F 30 Audi Max
DTSTART;TZID=Europe/Stockholm:20240604T095700
DTEND;TZID=Europe/Stockholm:20240604T095800
UID:submissions.pasc-conference.org_PASC24_sess158_pos101@linklings.com
SUMMARY:P43 - Quo Vadis: Helping Applications Manage On-Node Resources on 
 Modern Systems
DESCRIPTION:Poster\n\nEdgar A. Leon (Lawrence Livermore National Laborator
 y) and Samuel K. Gutierrez (Los Alamos National Laboratory)\n\nScientific 
 discovery is increasingly enabled by heterogeneous\nhardware that includes
  multiple processor types, deep memory\nhierarchies, and heterogeneous mem
 ories. To effectively utilize this\nhardware, computational scientists mus
 t compose their applications\nusing a combination of programming models, m
 iddleware, and runtime\nsystems. Since these systems are often designed in
  \nisolation from each other, their concurrent execution often results in\
 nresource contention and interference, which limits application\nperforman
 ce and scalability. This problem adds to the\nalready complex interactions
  between multiple physics libraries and\nemerging machine learning compone
 nts in scientific\napplications. Consequently, real-world applications fac
 e numerous\nchallenges on heterogeneous machines.\nThis poster presents Qu
 o Vadis, an interface and runtime system\nthat helps hybrid applications m
 ake efficient use of heterogeneous\nhardware, ease programmability in the 
 presence of multiple programming\nabstractions, and enable portability acr
 oss systems. The runtime\nsystem abstracts out low-level details of the ha
 rdware and presents an\narchitecture-independent interface applications ca
 n use to leverage\nlocal resources automatically and without user interven
 tion. The\nposter also includes a skeleton multi-physics application where
  we\napplied Quo Vadis to demonstrate how the challenges described above\n
 can be met in a portable way across systems and with a small effort\nfrom 
 application writers.\n\nSession Chair: Iva Kavcic (Met Office)
END:VEVENT
END:VCALENDAR
