high performance computing on graphics processing units: hgpu.org

The paper is devoted to creating effective and flexible Tsunami Modeling Environment based on a Service-Oriented Architecture (SOA) allowing high-level of operability and reusability of system components. Accordingly, we use the original Virtual MVC-design pattern (VMVC-pattern) approach that is demarcating a Functional (View) and an Implementation (Model) task by inducing an Integrator (Controller). This allows integrating a big variety of methods and services for Tsunami Modeling with respect to the various development platforms and architectures. The Model is organized on a set of layers in the form of Application Engines each of which is a subset of endpoint services that holds specific logic associated to a business process. Hence, an Engine can encompass functionalities of an API and realize processing that is specific to an application. The paper describes basic features of the MOST (Method of Splitting Tsunami) software package that was used as an initial Propagation Software Engine. This package was accepted by the USA National Ocean and Atmosphere Administration as the basic tool to calculate tsunami wave propagation and to create inundation maps. We describe a set of engines that was designed for several programming platforms including OpenMP, CELL architecture, and GPU’s allowing the flexible usage of available computational resources. Paper also includes an analysis the initial and output tsunami data, code design techniques as well as results of some numerical experiments and validation procedures.