Titlebook: High Performance Computing; 8th CCF Conference, Yunquan Zhang,Kenli Li,Zheng Xiao Conference proceedings 2013 Springer-Verlag Berlin Heide

蒸發(fā)

線

browbeat

An Improvement to the OpenMP Version of BoomerAMG,rtial differential equations. Parallel AMG has been widely used in large-scale scientific and engineering computation. In this paper, considering a class of linear algebraic equations with sparse and banded coefficient matrices, we improve the OpenMP version of BoomerAMG by modifying its modules of

堅(jiān)毅

Dynamic Partitioning of Scalable Cache Memory for SMT Architectures,future large-scale SMT processor. Although the design has good scalability, large-scale SMT architecture exacerbates the stress on cache, especially for the bank-interleaved data cache referred to in paper [1]. This paper proposes a dynamic partitioning method of scalable cache for large-scale SMT a

狂亂

淡紫色花

不知疲倦

Pathogen

The Design and Prototype Implementation of a Pipelined Heterogeneous Multi-core GPU,s processing unit, the HMGPU-9. HMGPU-9 supports OpenGL2.0 and DirectDraw with programmable vertex shaders and fragment shaders. It integrates 9 heterogeneous processor cores and many sophisticated application-specific accelerators into a XC6VLX550T FPGA. It employs dual-rail handshake protocol in i

Influx

不規(guī)則的跳動(dòng)