High-performance SIMT code generation in an active visual effects library

High-performance SIMT code generation in an active visual effects library
复制标题

主动视觉效果库中的高性能 SIMT 代码生成

DOI:
10.1145/1531743.1531772
复制
发表时间:
2009
期刊:
--
影响因子:
--
通讯作者:
Cornwall J
Cornwall J
中科院分区:
--
文献类型:
--
作者:
Cornwall J

文献摘要

参考文献

被引文献

相似文献

SIMT (Single-Instruction Multiple-Thread) is an emerging programming paradigm for high-performance computational accelerators, pioneered in current and next generation GPUs and hybrid CPUs. We present a domain-specific active-library supported approach to SIMT code generation and optimisation in the field of visual effects. Our approach uses high-level metadata and runtime context to guide and to ensure the correctness of optimisation-driven code transformations and to implement runtime-context-sensitive optimisations. Our advanced optimisations require no analysis of the original C++ kernel code and deliver 1.3x to 6.6x speedups over syntax-directed translation on GeForce 8800 GTX and GTX 260 GPUs with two commercial visual effects.
通过减少内存来增强数据局部性
DOI: --
发表时间: 2001
期刊: International Conference on Supercomputing
影响因子: --
作者:
Yonghong Song;Rong;Cheng Wang;Zhiyuan Li
通讯作者: Zhiyuan Li
主动视觉效果库中的显式依赖元数据
DOI: --
发表时间: 2007
期刊: International Workshop on Languages and Compilers for Parallel Computing
影响因子: --
作者:
Jay L. T. Cornwall;P. Kelly;Phil Parsonage;Bruno Nicoletti
通讯作者: Bruno Nicoletti
用于并行图像处理的骨架:SKIPPER 项目概述
DOI: --
发表时间: 2002
期刊: Parallel Computing
影响因子: 1.4
作者:
J. Sérot;D. Ginhac
通讯作者: D. Ginhac