In the field of structured parallel programming we study and implement a shared-memory runtime support for stencil data-parallel applications expressed according to the virtual processor formalism. It is a novel approach to introduce some optimizations that cannot be provided at compile time. The effectiveness of our runtime support is then evaluated with respect to optimized hand-written data-parallel applications
Stencil computations are an integral component of applications in a number of scientific computing d...
Application codes reliably achieve performance far less than the advertised capabilities of existing...
In this work, we present Dido, an implicitly parallel domain-specific language (DSL) that captures h...
The emergence of commercial multiprocessors has prompted computer scientists to take a closer look a...
Communicated by Guest Editors The implementation of stencil computations on modern, massively parall...
Contemporary state-of-the-art runtime systems underlying widely utilized general purpose parallel pr...
Application codes reliably achieve performance far less than the advertised capabilities of existing...
PhD ThesisAdvances in hardware technologies combined with decreased costs have started a trend towa...
Abstract Performance optimization of stencil computations has beenwidely studied in the literature, ...
While parallel programming is needed to solve large-scale scientific applications, it is more diffic...
This dissertation addresses creating portable and efficient parallel programs for scientific computi...
Advances in hardware technologies combined with decreased costs have started a trend towards massive...
Communicated by Guest Editors Our aim is to apply program transformations to stencil codes in order ...
The implementation of stencil computations on modern, mas-sively parallel systems with GPUs and othe...
The optimization of data parallel programs is a challenging open problem. We analyzed in detail the ...
Stencil computations are an integral component of applications in a number of scientific computing d...
Application codes reliably achieve performance far less than the advertised capabilities of existing...
In this work, we present Dido, an implicitly parallel domain-specific language (DSL) that captures h...
The emergence of commercial multiprocessors has prompted computer scientists to take a closer look a...
Communicated by Guest Editors The implementation of stencil computations on modern, massively parall...
Contemporary state-of-the-art runtime systems underlying widely utilized general purpose parallel pr...
Application codes reliably achieve performance far less than the advertised capabilities of existing...
PhD ThesisAdvances in hardware technologies combined with decreased costs have started a trend towa...
Abstract Performance optimization of stencil computations has beenwidely studied in the literature, ...
While parallel programming is needed to solve large-scale scientific applications, it is more diffic...
This dissertation addresses creating portable and efficient parallel programs for scientific computi...
Advances in hardware technologies combined with decreased costs have started a trend towards massive...
Communicated by Guest Editors Our aim is to apply program transformations to stencil codes in order ...
The implementation of stencil computations on modern, mas-sively parallel systems with GPUs and othe...
The optimization of data parallel programs is a challenging open problem. We analyzed in detail the ...
Stencil computations are an integral component of applications in a number of scientific computing d...
Application codes reliably achieve performance far less than the advertised capabilities of existing...
In this work, we present Dido, an implicitly parallel domain-specific language (DSL) that captures h...