A parallel rendezvous algorithm for interpolation between multiple grids

TL;DR: A majority of multi-physics-coupled applications concerns a situation where different physical phenomena occur locally in space, and a flexible integration infrastructure in which several independent solvers can be easily coupled with one another is more advantageous than to newly implement all the necessary functionalities and physical/numerical models onto a single code.

TL;DR: This work describes how rendezvous algorithms work in a scientific computing context and gives specific examples for molecular dynamics and Direct Simulation Monte Carlo codes which result in dramatic performance improvements versus simpler algorithms which do not scale as well.

TL;DR: A comparison of the three most frequently used multi-physics coupling strategies: the operator splitting, Picard iteration, and Jacobian-Free Newton–Krylov methods for virtual reactor coupling simulation.

TL;DR: This paper redesigns the algorithm in such a way that it maintains the scalability of the original CFD code on static grids and proposes a deterministic geometric decomposition on an intermediate “rendezvous” set of processes.

TL;DR: This work uses a binary decomposition of the domain to partition it into rectangles requiring equal computational effort, and studies the communication costs of mapping this partitioning onto different multiprocessors: a mesh- connected array, a tree machine, and a hypercube.

TL;DR: The concepts of workflow and production cycle open up a new horizon for integrating systems and software into what is called a distributed 'Teraflop-workbench' approach.

TL;DR: The major classes of algorithms are reviewed and their relative merits on problems from computational mechanics are discussed and suggestions are made for future research directions.

TL;DR: This paper considers parallelization of the smoothe aggregation multigrid methods, and discusses three different parallel aggregation algorithms an illustrates the advantages an disadvantages of each variant in terms of parallelism an convergence.