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:20210916T132447Z
LOCATION:Ernesto Bertarelli
DTSTART;TZID=Europe/Stockholm:20210709T110000
DTEND;TZID=Europe/Stockholm:20210709T113000
UID:submissions.pasc-conference.org_PASC21_sess189_pap110@linklings.com
SUMMARY:A Task-Based Distributed Parallel Sparsified Nested Dissection Alg
 orithm
DESCRIPTION:Paper\n\nA Task-Based Distributed Parallel Sparsified Nested D
 issection Algorithm\n\nCambier, Darve\n\nSparsified nested dissection (spa
 ND) is a fast scalable linear solver for sparse linear systems. It combine
 s nested dissection and separator sparsification, leading to an algorithm 
 with an O(N log N) complexity on many problems. In this work, we study the
  parallelization of spaND using TaskTorrent, a lightweight, distributed, t
 ask-based runtime in C++. This leads to a distributed version of spaND usi
 ng a task-based runtime system. We explain how to adapt spaND's partitioni
 ng for parallel execution, how to increase concurrency using a simultaneou
 s sparsification algorithm, and how to express the DAG using TaskTorrent. 
 We then benchmark spaND on a few large problems. spaND exhibits good stron
 g and weak scalings, efficiently using up to 9,000 cores when ranks grow s
 lowly with the problem size.\n\nDomain: CS and Math
END:VEVENT
END:VCALENDAR