Massive parallel computing

GPGPUs in computational finance: massive parallel computing for American style options

The pricing of American style and multiple exercise options is a very challenging problem in mathematical finance. One usually employs a Least-Square Monte Carlo approach (Longstaff-Schwartz method) for the evaluation of conditional expectations which arise in the Backward Dynamic Programming principle for such optimal stopping or stochastic control problems in a Markovian framework. Unfortunat...

Parallel Computing in Paderborn: The SFB 376 "Massive Parallelism - Algorithms, Design Methods, Applications"

Multi-domain large deformations by monolithic approach for massive parallel computing

A monolithic approach is presented to solve multi-domain large deformations problems. It is based on an eulerian formulation using a fixed yet anisotropic adaptative mesh. Within this context, a level set method is used to capture the interface evolution of different domains. In terms of parallel computing, it is expected that the technique of single mesh and single mechanical solver developed ...

High Performance Computing: Grid Computing & Parallel Programming

Science is constantly pushing the limits of computational and storage power and even the fastest supercomputers are not capable of satisfying the computational demand. Recent advances in hardware technologies have provided more computing power than ever before and with the advent of multiple core processors we can expect this trend to continue. It is expected that processors with hundreds of co...

General Purpose Parallel Computing

A major challenge for computer science in the 1990s is to determine the extent to which general purpose parallel computing can be achieved. The goal is to deliver both scalable parallel performance and architecture independent parallel software. (Work in the 1980s having shown that either of these alone can be achieved.) Success in this endeavour would permit the long overdue separation of soft...