Various servers with different characteristics and architectures are hitting the market, and their evaluation and comparison in terms of HPC features is complex and multidimensional. In this paper, we share our experience of evaluating a diverse set of HPC systems, consisting of three mainstream and five emerging architectures. We evaluate the performance and power efficiency using prominent HPC benchmarks, High-Performance Linpack (HPL) and High Performance Conjugate Gradients (HPCG), and expand our analysis using publicly available specialized kernel benchmarks, targeting specific system components. In addition to a large body of quantitative results, we emphasize six usually overlooked aspects of the HPC platforms evaluation, and share our conclusions and lessons learned. Overall, we believe that this paper will improve the evaluation and comparison of HPC platforms, making a first step towards a more reliable and uniform methodology.

This paper presents a pipeline for high quality volume rendering of adaptive mesh refinement (AMR) datasets. We introduce a new method allowing high quality visualization of hexahedral cells in this context; this method avoids artifacts like discontinuities in the isosurfaces. To achieve this, we choose the number and placement of sampling points over the cast rays according to the analytical properties of the reconstructed signal inside each cell. We extend our method to handle volume shading of such cells. We propose an interpolation scheme that guarantees continuity between adjacent cells of different AMR levels. We introduce an efficient hybrid CPU-GPU mesh traversal technique. We present an implementation of our AMR visualization method on current graphics hardware, and show results demonstrating both the quality and performance of our method.