Titlebook: Sustained Simulation Performance 2015; Proceedings of the j Michael M. Resch,Wolfgang Bez,Sabine Roller Conference proceedings 2015 Springe

Liberate

A Data Management Policy for Energy-Efficient Cache Mechanismsvicted from cache memories. Since these blocks do not contribute to performance improvement, they limit the energy saving achieved by dynamic cache resizing mechanisms, which will play key roles to improve the energy efficiency of future microprocessors. The evaluation results show that, by eliminat

engender

Toward Flexible Supercomputing and Visualization Systemeed networks for research and education. Especially in the era where the proliferation of scientific data to analyze is accelerating, limited and finite computational and network resources are to be efficiently and flexibly leveraged and shared among scientists. Also, computing requirements for supe

overbearing

NEC Vector Supercomputer: Its Present and Futureemory-intensive scientific applications. The major concept of SX-ACE is the provision of a world top-level single core performance of 64 GFlop/s, as well as the world largest memory bandwidth per single core of 256 GB/s in its maximum with a high power efficiency. It is also designed to make availab

constitutional

pessimism

HERTZ

A High-Level Interface of Xevolver for Composing Loop Transformationsework originally assumes that transformation rules are written in XSLT, which might be too low-level for standard users to express their own rules. Hence, this article describes a high-level interface of the Xevolver framework, which offers an easy and declarative way to define a custom compiler dir

星球的光亮度

Performance Evaluation of Compiler-Assisted OpenMP Codes on Various HPC Systemsunderstand its code structure and data dependency, resulting in effective automatic optimizations. However, these code modifications might not be effective for a different compiler because the different compiler cannot always parallelize the modified code. In this paper, in order to achieve effectiv

不可磨灭

Code Optimization Activities Toward a High Sustained Simulation Performanceo the increases in system complexity. To create significant scientific and innovative results by numerical simulations using supercomputers, the collaboration of computational scientists and computer scientists is mandatory. In this article, joint research activities and the code optimization suppor

OGLE

The Multicore Challenge: Petascale DNS of a Spatially-Developing Supersonic Turbulent Boundary Layertigation of wall-bounded turbulent flows. The applied numerical method, however, has to enable an efficient usage of high performance computing systems to satisfy the involved computational costs. In this context, discontinuous Galerkin (DG) methods have become a promising candidate to conduct DNS i

Esophagus

High-Resolution Latitude Belt Simulation with the Weather Research and Forecasting Modelolution can be too coarse to represent certain meteorological features, often limited area models (LAM) on higher resolutions are applied for the region of interest. They require external boundary conditions from a coarser driving model at the edges of the model domain. As this can deteriorate the r