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:20220812T074357Z
LOCATION:Samarkand Room
DTSTART;TZID=Europe/Stockholm:20220629T110000
DTEND;TZID=Europe/Stockholm:20220629T130000
UID:submissions.pasc-conference.org_PASC22_sess157@linklings.com
SUMMARY:MS5B - Numerical Methods for Solving Large Scale Linear Systems on
  Modern Exascale Computers for Industrial Applications
DESCRIPTION:Minisymposium\n\nNumerical simulation is a very common tool in
  a wide range of applications, ranging from structural mechanics to comput
 ational fluid-dynamics, underground processes, electromagnetism, and many 
 others. Nowadays, the industrial world is constantly pushing for increasin
 gly large and accurate, real-time solutions and modern engineers need effi
 cient tools to address numerical simulations that require the solution of 
 sparse linear systems of millions or even billions of unknowns. A common b
 ottleneck in large size simulations is the cost for solving a linear syste
 ms of equations which tend to grow superlinearly with the problem size. Th
 us, the use of High-Performance Computing (HPC) is increasingly necessary,
  and the development of efficient and scalable sparse linear solvers is an
  important research topic. From a hardware point of view, the trend in HPC
  is towards massively parallel computing through smart interconnected node
 s equipped with accelerators featuring thousands of simple computing units
  (cores). The availability of these novel computational platforms boosts b
 oth a theoretical renovation of well-established algorithms and the develo
 pment of brand-new ideas. The algorithms must be designed and optimized ha
 ving as target platforms multiple nodes with thousands of simple computing
  units and a memory hierarchy that is much more exposed to the developer's
  control.\n\nThe Use of Filtering and Energy Minimization to Accelerate AM
 G in Large Size Mechanical Problems\n\nCipolletta, Janna, Spiezia, Rondina
 \n\nMultigrid methods are very well-known tools in scientific computing be
 cause of their capabilities in solving linear systems with nearly optimal 
 convergence rates. In particular, in the field of computational mechanics,
  where one needs to solve large linear systems represented by sparse matri
 ces, the...\n\n---------------------\nA GPU-Based, Industrial-Grade Compos
 itional Reservoir Simulator\n\nFerrari\n\nRecently, graphics processing un
 its (GPUs) have been demonstrated to provide a significant performance ben
 efit for industrial applications where the computational bottleneck is the
  numerical solution of partial differential equations. In this work we sho
 w how GPUs can be effectively used for complex...\n\n---------------------
 \nRecent Advances in Randomized Techniques for Linear Algebra\n\nGrigori, 
 Balabanov\n\nThis talk will review several recent advances in using random
 ization for large scale linear algebra operations. We will focus in partic
 ular on solving linear systems by using Krylov subspace methods. In this c
 ontext we will introduce a randomization techniques for orthogonalizing a 
 set of vectors an...\n\n---------------------\nOn the Design of Algebraic 
 Multigrid Methods for Exascale Computers and their Use in Numerical Simula
 tions\n\nYang\n\nWith the increasing inclusion of accelerators into curren
 t and future high-performance computers, it has become important to enable
  parallel mathematical libraries to take advantage of the increased perfor
 mance potential of GPUs. The hypre software library provides high performa
 nce preconditioners a...\n\n\nDomain: Computer Science and Applied Mathema
 tics, Engineering, Physics
END:VEVENT
END:VCALENDAR