CUDA / OpenCL learning sources?

agolangf · 2015-10-05 11:16:07 · 2429 次点击    
这是一个分享于 2015-10-05 11:16:07 的资源,其中的信息可能已经有所发展或是发生改变。

So I've been trying to find a good book for learning CUDA and OpenCL implementation with Go, or even a PDF, or a video. But all i see to find is the same article for CUDA, which is more of a demonstration, and even worse with OpenCL where nothing is avaliable (at least that I have been able to find..).

Now of course the packages functions are correctly documentated, as usual. But it isn't of much usefulness without a proper guide or introduction to it.

Any sources, if there are any at all, would be much appreciated :smile:


评论:

mwholt:

I've never done CUDA in Go (would be surprised if it was worth it) but here's an awesome guide: https://github.com/romain-jacotin/cuda

DarkAnHell:

Awesome, seems good at a first glance. Thanks for sharing, I will definetly spend some time with it.

Also, still nothing about OpenCL (?). Seems weird to me that no one bothered writing something about it

picardo:

Not a book, but there is a course on Coursera called Heterogeneous Parallel Programming that covers CUDA.

DarkAnHell:

Will take a look, thanks :)

space-llama:

I've tried to go down this route for a project that was supposed to run on a dozen GPU-heavy machines, but found the community support for CUDA via Go kinda lacking. Basically all of the concerns I had about the right way to work with CUDA were answered by specific examples in official docs, not GitHub Issues for projects. Try the C++ libraries, especially Thrust, it's soooooo much easier (http://docs.nvidia.com/cuda/thrust/) and more...appropriate?

DarkAnHell:

Oh well, I guess i'm out of luck with Go here. Thanks for the recomendation, i will keep it in mind


入群交流(和以上内容无关):加入Go大咖交流群,或添加微信:liuxiaoyan-s 备注:入群;或加QQ群:692541889

2429 次点击  
加入收藏 微博
0 回复
暂无回复
添加一条新回复 (您需要 登录 后才能回复 没有账号 ?)
  • 请尽量让自己的回复能够对别人有帮助
  • 支持 Markdown 格式, **粗体**、~~删除线~~、`单行代码`
  • 支持 @ 本站用户;支持表情(输入 : 提示),见 Emoji cheat sheet
  • 图片支持拖拽、截图粘贴等方式上传