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:20220812T074335Z
LOCATION:Osaka Room
DTSTART;TZID=Europe/Stockholm:20220629T110000
DTEND;TZID=Europe/Stockholm:20220629T113000
UID:submissions.pasc-conference.org_PASC22_sess167_msa213@linklings.com
SUMMARY:VecMem: A Library for Managing Heterogeneous Application Memory Us
 ing Modern C++
DESCRIPTION:Minisymposium\n\nVecMem: A Library for Managing Heterogeneous 
 Application Memory Using Modern C++\n\nKrasznahorkay, Swatman, Gessinger\n
 \nUsing modern, high-level C++ code in heterogeneous programming is made d
 ifficult by the extra dimensions of phase space used in managing the memor
 y of the code. In order to use the memory of non-CPU devices, one generall
 y needs to make use of allocation/deallocation functions fitting a C appli
 cation's programming style. In this talk we present vecmem, a library aimi
 ng to make it easier to use modern C++ code on accelerators, leveraging th
 e functionality introduced with C++17's memory resources. The library prov
 ides a composable set of memory resource types that allow using the memory
  of accelerators through the CUDA, HIP and SYCL runtime libraries using st
 andard C++ containers, and types mimicking standard containers for use in 
 device code. We will show how this project makes it easier to write effici
 ent and safe code for managing the memory of heterogeneous applications, a
 nd what performance benefits we can achieve through clever memory manageme
 nt.\n\nDomain: Computer Science and Applied Mathematics, Physics
END:VEVENT
END:VCALENDAR
