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:20260605T154541Z
LOCATION:Bldg. 6 - Room 002
DTSTART;TZID=Europe/Stockholm:20260629T160000
DTEND;TZID=Europe/Stockholm:20260629T163000
UID:submissions.pasc-conference.org_PASC26_sess114_msa145@linklings.com
SUMMARY:Dense Numerical Linear Algebra (NLA) Under the Microscope: What Sc
 ientific Applications Really Need
DESCRIPTION:Xinzhe Wu (Jülich Supercomputing Centre) and Edoardo Di Napoli
  (Forschungszentrum Jülich)\n\nDense numerical linear algebra (NLA) is a f
 undamental building block in many scientific applications, supporting simu
 lations, data analysis, and modeling workflows across diverse domains. Whi
 le significant advances have been made in algorithms and high-performance 
 implementations, the evolution of NLA libraries is often driven by theoret
 ical developments and hardware trends rather than systematic feedback from
  application developers.\n\nIn this talk, we present insights from an appl
 ication-oriented survey conducted within ExaNLA, aimed at understanding ho
 w dense NLA functionality is used in real scientific codes. The survey col
 lected input from developers of a broad range of applications regarding th
 e dense linear algebra routines they rely on, how these routines are integ
 rated into their workflows, and which limitations they encounter when usin
 g existing libraries.\n\nThe results reveal common patterns across domains
 , including a strong reliance on a relatively small set of dense operation
 s, as well as emerging needs related to scalability, flexibility, and hete
 rogeneous architectures. These findings highlight gaps between current NLA
  capabilities and the evolving requirements of scientific software, provid
 ing an application-driven perspective for future NLA developments.\n\nDoma
 in: Chemistry and Materials, Climate, Weather, and Earth Sciences, Computa
 tional Methods and Applied Mathematics\n\nSession Chair: Edoardo Di Napoli
  (Forschungszentrum Jülich)\n\n
END:VEVENT
END:VCALENDAR