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:20220812T074334Z
LOCATION:Foyer 2nd Floor
DTSTART;TZID=Europe/Stockholm:20220628T090000
DTEND;TZID=Europe/Stockholm:20220628T110000
UID:submissions.pasc-conference.org_PASC22_sess181_pos141@linklings.com
SUMMARY:P40 - Towards a Task Based GPU Enabled Distributed Eigenvalue Solv
 er
DESCRIPTION:Poster\n\nP40 - Towards a Task Based GPU Enabled Distributed E
 igenvalue Solver\n\nInvernizzi, Nikolov, Reverdell, Simberg, Solcà\n\nDeve
 loping and implementing an efficient GPU enabled eigenvalue solver is a co
 mplex operation, which becomes a challenge when a task-based approach is u
 sed. A fine balance among the number of tasks and their execution time has
  to be found, to maintain enough parallelism and to avoid increasing the s
 cheduler overhead.<br />However the benefits of task-based linear algebra 
 implementations are important, as our previous work on distributed Cholesk
 y decomposition and triangular solver has shown.<br />The reduction of the
  number of synchronisation points (compared to the fork join approach used
  by LAPACK and ScaLAPACK) and the possibility of scheduling multiple algor
 ithms to run concurrently are two of the main benefits.<br />Here we prese
 nt a task-based GPU-enabled eigenvalue solver based on the pika library. P
 ika was chosen because it follows the developments of Concurrency and Para
 llelism proposed as part of the ongoing C++ standardization process. In pa
 rticular pika provides an implementation of the latest sender/receiver pro
 posal.
END:VEVENT
END:VCALENDAR