请选择 进入手机版 | 继续访问电脑版

开源计算机图形学社区(Open Source Computer Graphics Community) |OpenGPU Forum (2007-2013)| OpenGPU Project

 找回密码
 注册
搜索
查看: 5103|回复: 7

PhD 论文: Scientific Computing on Heterogeneous Architectures [复制链接]

Rank: 20Rank: 20Rank: 20Rank: 20Rank: 20

注册时间
2010-4-11
积分
1118

会员荣誉勋章

发表于 2011-2-12 04:39:00 |显示全部楼层
写的还不错,第二章的current paper review不错。值得一看,不过总体来说heterogeneous system的问题还是memory system, 目前一个趋势是semi-automatic local memory management。


PhD Thesis: Scientific Computing on Heterogeneous ArchitecturesDecember 22nd, 2010

Abstract:

The CPU has traditionally been the computational work horse in scientific computing, but we have seen a tremendous increase in the use of accelerators, such as Graphics Processing Units (GPUs), in the last decade. These architectures are used because they consume less power and offer higher performance than equivalent CPU solutions. They are typically also far less expensive, as more CPUs, and even clusters, are required to match their performance. Even though these accelerators are powerful in terms of floating point operations per second, they are considerably more primitive in terms of capabilities. For example, they cannot even open a file on disk without the use of the CPU. Thus, most applications can benefit from using accelerators to perform heavy computation, whilst running complex tasks on the CPU. This use of different compute resources is often referred to as heterogeneous computing, and we explore the use of heterogeneous architectures for scientific computing in this thesis. Through six papers, we present qualitative and quantitative comparisons of different heterogeneous architectures, the use of GPUs to accelerate linear algebra operations in MATLAB, and efficient shallow water simulation on GPUs. Our results show that the use of heterogeneous architectures can give large performance gains.

(André R. Brodtkorb, “Scientific Computing on Heterogeneous Architectures”, Ph.D. thesis, University of Oslo, Faculty of Mathematics and Natural Sciences, 2010)

原地址



http://gpgpu.org/2010/12/22/phd-thesis-brodtkorb
1

查看全部评分

头像被屏蔽

版主

Missionary

Rank: 20Rank: 20Rank: 20Rank: 20Rank: 20

注册时间
2010-4-25
积分
10777
发表于 2011-2-12 10:45:44 |显示全部楼层
提示: 作者被禁止或删除 内容自动屏蔽
签名被屏蔽

使用道具 举报

Rank: 20Rank: 20Rank: 20Rank: 20Rank: 20

注册时间
2010-4-11
积分
1118

会员荣誉勋章

发表于 2011-2-14 09:40:11 |显示全部楼层
回复 zenny_chen 的帖子

呵呵,好久没上opengpu了,回来发现论坛系统改进很大. 很给力~

使用道具 举报

Rank: 20Rank: 20Rank: 20Rank: 20Rank: 20

注册时间
2011-4-19
积分
1754
发表于 2011-4-19 18:01:44 |显示全部楼层
总觉得异构系统没法解决NUMA式构架对内存访问的控制,即便是把GPU和CPU做到一块板子上,这个问题依然存在。。。

使用道具 举报

Rank: 20Rank: 20Rank: 20Rank: 20Rank: 20

注册时间
2011-5-11
积分
1297
发表于 2011-5-23 08:09:38 |显示全部楼层
恩,这篇文章好,呵呵,正准备写一篇关于OpenCL的综述

使用道具 举报

Rank: 5Rank: 5

注册时间
2011-1-17
积分
96
发表于 2011-7-28 08:17:55 |显示全部楼层
回复 jerry_gzy 的帖子

Good! Thanks for sharing!

使用道具 举报

Rank: 8Rank: 8

注册时间
2011-3-14
积分
124
发表于 2011-8-30 06:33:34 |显示全部楼层
Thanks for sharing

使用道具 举报

Rank: 4

注册时间
2016-2-25
积分
22
发表于 2016-12-28 13:13:07 |显示全部楼层
牛文,阅读以下

使用道具 举报

最近看过此主题的会员

您需要登录后才可以回帖 登录 | 注册

‹‹
我的工具栏

关于我们|手机版|Archiver|开源计算机图形学社区(Open Source Computer Graphics Community) | OpenGPU Project | OpenGPU Forum (2007-2013)

GMT+8, 2017-2-26 20:14 , Processed in 0.082081 second(s), 14 queries .

Powered by Discuz! X2

© 2001-2011 Comsenz Inc.

回顶部