Multicore processors have become the dominant industry trend to increase computer systems performance, driving electromagnetics (EM) practitioners to redesign their applications using parallel programming paradigms. This is especially true for computations involving complex data structures such as sparse matrix computations that often arise in EM simulations with the finite element method (FEM). These computations require pointer manipulation that render useless many compiler optimizations and parallel shared memory frameworks (e.g. OpenMP). This work presents new sparse data structures and techniques to efficiently exploit multicore parallelism and short-vector units (the last of which has not been exploited by state of the art sparse matr...
We contribute to the optimization of the sparse matrix-vector product by introducing a variant of th...
Electromagnetic field computation with the Finite Element (FE) method implies solving of large line...
This paper presents a novel parallelization approach to speedup EMT simulation, using GPU-based comp...
A wide class of finite-element (FE) electromagnetic applications requires computing very large spars...
International audienceNowadays, several industrial applications are being ported to parallel archite...
We are witnessing a dramatic change in computer architecture due to the multicore paradigm shift, as...
The solution of large and complex electromagnetic (EM) problems often leads to a substantial demand ...
We are witnessing a dramatic change in computer architecture due to the multicore paradigm shift, as...
In this paper, the performance of a parallel sparse direct solver on a shared memory multicore syste...
The Finite Element Method (FEM) is a computationally intensive scientific and engineering analysis t...
AbstractThe sparse matrix-vector multiplication (SpMV) is a fundamental kernel used in computational...
Multiphysics simulations are at the core of modern Computer Aided Engineering (CAE) allowing the ana...
Sparse computations are ubiquitous in computational codes, with the sparse matrix-vector (SpMV) mult...
This dissertation presents an architecture to accelerate sparse matrix linear algebra,which is among...
In this work, the relation between the sparsity patterns of sparse matrices created through FEM desc...
We contribute to the optimization of the sparse matrix-vector product by introducing a variant of th...
Electromagnetic field computation with the Finite Element (FE) method implies solving of large line...
This paper presents a novel parallelization approach to speedup EMT simulation, using GPU-based comp...
A wide class of finite-element (FE) electromagnetic applications requires computing very large spars...
International audienceNowadays, several industrial applications are being ported to parallel archite...
We are witnessing a dramatic change in computer architecture due to the multicore paradigm shift, as...
The solution of large and complex electromagnetic (EM) problems often leads to a substantial demand ...
We are witnessing a dramatic change in computer architecture due to the multicore paradigm shift, as...
In this paper, the performance of a parallel sparse direct solver on a shared memory multicore syste...
The Finite Element Method (FEM) is a computationally intensive scientific and engineering analysis t...
AbstractThe sparse matrix-vector multiplication (SpMV) is a fundamental kernel used in computational...
Multiphysics simulations are at the core of modern Computer Aided Engineering (CAE) allowing the ana...
Sparse computations are ubiquitous in computational codes, with the sparse matrix-vector (SpMV) mult...
This dissertation presents an architecture to accelerate sparse matrix linear algebra,which is among...
In this work, the relation between the sparsity patterns of sparse matrices created through FEM desc...
We contribute to the optimization of the sparse matrix-vector product by introducing a variant of th...
Electromagnetic field computation with the Finite Element (FE) method implies solving of large line...
This paper presents a novel parallelization approach to speedup EMT simulation, using GPU-based comp...