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:20191206T142453Z
LOCATION:Substanz 1\, 2
DTSTART;TZID=Europe/Stockholm:20180626T083000
DTEND;TZID=Europe/Stockholm:20180626T100000
UID:isc_hpc_ISC High Performance 2018_sess113_post122@linklings.com
SUMMARY:(RP11) Task-Parallel Domain-wise Multifrontal Solver for High-Perf
 ormance Finite Element Analysis on Multi-core Systems
DESCRIPTION:Research Poster\n\n(RP11) Task-Parallel Domain-wise Multifront
 al Solver for High-Performance Finite Element Analysis on Multi-core Syste
 ms\n\nKim, Lee, Cho, Park, Kim\n\nThe domain-wise multifrontal solver (DMF
 S) is a high-performance sparse direct solver especially designed for effi
 cient finite element analysis and it has many successful applications to c
 ommercial and research finite element analysis codes. However, the DMFS ha
 s limited scalability due to simple static scheduling based on the elimina
 tion tree constructed by nested dissection for matrix ordering. In this re
 search, the task-parallel DMFS for high-performance finite element analysi
 s on multi-core systems is developed by using Thread Building Block (TBB) 
 to implement task parallelism for efficient dynamic scheduling. Task paral
 lelism in DMFS is realized just by spawning tasks with consideration for t
 he size and dependency of each task. Front nodes within appropriate number
  of levels in the elimination tree are spawned as tasks. Partial factoriza
 tion for each spawned front node also spawns tasks for computations for bl
 ocked matrices. All spawned tasks are dynamically scheduled by TBB schedul
 er. By implementing task parallelism, the parallel performance of the DMFS
  is improved significantly compared with the original DMFS based on static
  scheduling. To evaluate the overall performance of finite element analysi
 s using the DMFS, several finite element models for structural analysis wi
 th more than millions of degrees of freedom are analyzed and computing tim
 es are compared with state-of-the-art finite element analysis codes and sp
 arse solvers. According to the performance comparison results, the perform
 ance of the task-parallel DMFS is much superior to state-of-the-art finite
  element analysis codes and sparse solvers.\n\nPasses: Conference Pass, Pa
 rallel Algorithms, Parallel Applications, Scientific Software Development\
 n\nTag: Conference Pass, Parallel Algorithms, Parallel Applications, Scien
 tific Software Development
URL:https://2018.isc-program.com/presentation/?id=post122&sess=sess113
END:VEVENT
END:VCALENDAR