Josef Weinbub

Learn More
Quality and size of mesh elements are important for optimizing the accuracy and convergence of mesh-based simulation processes. Often, a priori information, like internal material properties, of regions of interest is available, which can be used to locally specify the mesh element size for finding a good balance between the mesh resolution on the one hand(More)
We revisit the implementation of iterative solvers on discrete graphics processing units and demonstrate the benefit of implementations using extensive kernel fusion for pipelined formulations over conventional implementations of classical formulations. The proposed implementations with both CUDA and OpenCL are freely available in ViennaCL and are shown to(More)
—Creating multiple meshes of a semiconductor device by varying specific geometric properties, like the gate length of a MOSFET, is a crucial step for optimization or scaling processes of these devices. A geometry generation technique for semiconductor devices using geometry templates is presented and implemented in the open source meshing tool ViennaMesh,(More)
One of the major drawbacks of computing with graphics adapters is the limited available memory for relevant problem sizes. To overcome this limitation for the ViennaCL library, we investigate a partitioning approach for one of the standard benchmark problems in High-Performance Computing (HPC), namely the dense matrix-matrix product. We apply this(More)
The influence of multi-core central processing units and graphics processing units on several algebraic multigrid methods is investigated in this work. Different performance metrics traditionally employed for algebraic multigrid are reconsidered and reevaluated on these novel computing archi-tectures. Our benchmark results show that with the use of graphics(More)
—To provide simulation software in the field of TCAD with the utmost flexibility regarding generation and adaptation of meshes, a generic and high-quality meshing library, ViennaMesh, has been developed. The library is coded in C++ and utilizes modern programming techniques to wrap tasks, like mesh generation and mesh adaptation, into functional objects,(More)
The performance portability of OpenCL kernel implementations for common memory bandwidth limited linear algebra operations across different hardware generations of the same vendor as well as across vendors is studied. Certain combinations of kernel implementations and work sizes are found to exhibit good performance across compute kernels, hardware(More)
The Wigner Monte Carlo method, based on the generation and annihilation of particles, has emerged as a promising approach to treat transient problems of quantum electron transport in nanostructures. Tackling these simulations in multiple spatial dimensions demands a parallelized approach to facilitate a practical application of the method in order to(More)