BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Parallelization and Software Concepts for Tsunami Simulation on Dy
 namically Adaptive Triangular Grids - Bader\, M (Technische Universitt Mnc
 hen)
DTSTART:20120824T133000Z
DTEND:20120824T140000Z
UID:TALK39389@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:We present a memory- and cache-efficient approach for simulati
 ons on recursively refined dynamically adaptive triangular grids. Grid cel
 ls are stored and processed in an order defined by the Sierpinski curve\; 
 the resulting locality properties are exploited for optimised serial imple
 mentation and parallelisation. The approach is particularly designed for F
 inite Volume and discontinuous Galerkin solvers\, with Tsunami simulation 
 as the main target application. In the talk\, we will discuss approaches f
 or parallelisation in shared and distributed memory. We will present a cla
 ssical partitioning-based strategy\, as well as a novel shared-memory appr
 oach based on the dynamical scheduling of many small sub-partitions. Here\
 , the intention is to allow for strongly varying computational load per el
 ement (as required for inundation modelling\, or for local time-stepping m
 ethods). \n\nIn addition\, we would like to discuss some ideas on how to p
 rovide bathymetry data and (time-dependent) displacements for a simulation
  with dynamically adaptive refinement. We'll present some first results of
  adaptive simulations using the augmented Riemann solvers provided with Ge
 oClaw. \n\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR