Hierarchically Tiled Arrays (HTAs) are data structures that facilitate locality and parallelism of array intensive computations with block-recursive nature. The model underlying HTAs provides programmers with a global view of distributed data as well as a single-threaded view of the execution. In this paper we present htalib, a C++ implementation of HTAs. This library provides several novel con...

A leading approach to improving the accuracy on numerical relativity simulations of black hole systems is through fixed or adaptive mesh refinement techniques. We describe a generic numerical error which manifests as slowly converging, artificial reflections from refinement boundaries in a broad class of mesh-refinement implementations, potentially limiting the effectiveness of meshrefinement t...

We consider a postprocessor that is able to analyze the flow-field generated by an external (unknown) code so as to determine the error of useful functionals. The residuals generated by the action of a high order finite-difference stencil on a numerically computed flow-field are used for adjoint based a-posteriori error estimation. The method requires information on the physical model (PDE syst...

The possibilities to grow isolated structures of complex oxides by pulsed laser deposition through stencils were investigated. A stencil consisting of a SiN membrane with apertures of several hundred nanometers embedded in a Si chip is placed in front of a heated substrate (up to 750 degrees Celsius). Deposition through these apertures results in resistless, direct patterning by local depositio...

The OpenACC programming model simplifies the programming for accelerator devices such as GPUs. Its abstract accelerator model defines a least common denominator for accelerator devices, thus it cannot represent architectural specifics of these devices without losing portability. Therefore, this general-purpose approach delivers good performance on average, but it misses optimization opportuniti...