Search

The Overture framework is an object-oriented environment for solving partial differential equations in two and three space dimensions. It is a collection of C++ libraries that enables the use of finite difference and finite volume methods at a level that hides the details of the associated data structures. Overture can be used to solve problems in complicated, moving geometries using the method of overlapping grids. It has support for grid generation, difference operators, boundary conditions, data-base access and graphics. In this paper we briefly present Overture, present some of the newer grid generation capabilities, and discuss our approach toward performance within Overture and the A++P++ array class abstractions upon which Overture depends, this work represents some of the newest work in Overture. The results we present show that the abstractions represented within Overture and the A++P++ array class library can be used to obtain application codes with performance equivalent to that of optimized C and Fortran 77. Further, the preprocessor mechanism for which approach we present results, is general in its application to any object-oriented framework or application and is not specific to Overture. [ABSTRACT FROM AUTHOR]

Application codes reliably achieve performance far less than the advertised capabilities of existing architectures, and this problem is worsening with increasingly-parallel machines. For large-scale numerical applications, stencil operations often impose the greater part of the computational cost, and the primary sources of inefficiency are the costs of message passing and poor cache utilization. This paper proposes and demonstrates optimizations for stencil and stencil-like computations for both serial and parallel environments that ameliorate these sources of inefficiency. Additionally, we argue that when stencil-like computations are encoded at a high level using object-oriented parallel array class libraries, these optimizations, which are beyond the capability of compilers, may be automated. The automation of these optimizations is particularly important since the transformations represented by cache based optimizations can be unreasonably complicated by the peculiarities which are architecture specific. This paper briefly presents the approach toward the automation of these transformations. [ABSTRACT FROM AUTHOR]