BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160556Z
LOCATION:Track 1
DTSTART;TZID=America/New_York:20201111T114000
DTEND;TZID=America/New_York:20201111T120500
UID:submissions.supercomputing.org_SC20_sess193_ws_hipar102@linklings.com
SUMMARY:Introducing Multi-Level Parallelism, at Coarse, Fine, and Instruct
 ion Level to Enhance the Performance of Iterative Solvers for Large Sparse
  Linear Systems on Multi- and Many-Core Architecture
DESCRIPTION:Workshop\n\nIntroducing Multi-Level Parallelism, at Coarse, Fi
 ne, and Instruction Level to Enhance the Performance of Iterative Solvers 
 for Large Sparse Linear Systems on Multi- and Many-Core Architecture\n\nGr
 atien\n\nMulti-core and many-core systems are now a common feature of new 
 hardware architectures. The introduction of a very large number of cores a
 t the processor level requires multi-level parallelism to fully take advan
 tage of the offered computing power. The induced  programming effort can b
 e fixed with parallel programming models based on the data flow model and 
 the task programming paradigm.   Standard numerical algorithms must then b
 e revisited to be parallelized at the finest levels. Iterative linear solv
 ers are a key part of petroleum reservoir simulation representing up to 80
 % of the total computing time. Standard preconditioning methods for large,
  sparse matrices -- such as Incomplete LU Factorization (ILU) or Algebraic
  Multigrid (AMG) -- fail to scale on architectures with a large number of 
 cores. \n\nWe reconsider preconditioning algorithms to better introduce mu
 lti-level parallelism at both coarse level with MPI, at fine level with th
 reads, and at the instruction level to enable SIMD optimizations.  We enha
 nce the implementation of preconditioners like the multi-level domain deco
 mposition~(DDML) preconditioners, based on the popular Additive Schwartz M
 ethod (ASM), or the classical ILU0 preconditioner with the fine grained pa
 rallel fixed point variant. Our approach is validated on linear systems ex
 tracted from realistic petroleum reservoir simulations. The robustness of 
 the preconditioners is tested with respect to the data heterogeneities of 
 the study cases.  We evaluate the extensibility of our implementation rega
 rding the model sizes and its scalability regarding the large number of co
 res provided by KNL or SkyLake processors.\n\nTag: Extreme Scale Computing
 , Heterogeneous Systems, Parallel Programming Languages, Libraries, and Mo
 dels, Portability, Resource Management and Scheduling, Scalable Computing\
 n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR