任何优势MPI + CUDA超过只是纯粹的MPI？ [英] Any Advantage of MPI+CUDA over just pure MPI?

问题描述

通常的方法，以加快应用程序是使用MPI或更高级别的库，例如其中的PETSc使用MPI发动机罩下的并行应用程序。

The usual way to speed up an application is to parallelize an application using MPI or higher level libraries like PETSc which use MPI under the hood.

不过现在大家似乎有意利用CUDA并行化的应用程序或使用MPI和CUDA的混合的更加雄心勃勃的/大问题。

However nowadays everyone seems to be interested in using CUDA for parallelizing their application or using a hybrid of MPI and CUDA for more ambitious/larger problems.

是否有使用混合MPI + CUDA编程模型比传统的任何显着的优势，久经考验的并行编程MPI模式？我特别问这在粒子方法的应用领域

Is there any noticeable advantage in using a hybrid MPI+CUDA programming model over the traditional , tried and tested MPI model of parallel programming? I am asking this specifically in the application domains of particle methods

原因之一，我问这个问题是无处不在网络上我看到的声明说：粒子的方法自然地映射到GPU的体系结构或这方面的一些变化。但从来没有这样做，他们似乎证明，为什么我会用CUDA比使用MPI同样的工作更好。

One reason why I am asking this question is that everywhere on the web I see the statement that "Particle methods map naturally to the architecture of GPU's" or some variation of this. But never do they seem to justify why I would be better of using CUDA than using just MPI for the same job.

推荐答案

这是一个有点苹果和桔子。

This is a bit apples and oranges.

MPI与CUDA有着根本不同的体系结构。最重要的是，MPI可以让你在多个节点分发您的应用程序，而CUDA，您可以使用本地节点内的GPU。如果MPI程序的并行进程需要很长时间才能完成，那么，你应该看看他们如何能加快通过使用CPU的GPU，而不是做他们的工作。相反，如果你的CUDA应用程序仍然需要很长时间才能完成，你可能要分发的工作，采用MPI多节点。

MPI and CUDA are fundamentally different architectures. Most importantly, MPI lets you distribute your application over several nodes, while CUDA lets you use the GPU within the local node. If in an MPI program your parallel processes take too long to finish, then yes, you should look into how they could be sped up by using the GPU instead of the CPU to do their work. Conversely, if your CUDA application still takes too long to finish, you may want to distribute the work to multiple nodes using MPI.

这两种技术是pretty的多正交（假设所有节点的群集上支持CUDA的）。

The two technologies are pretty much orthogonal (assuming all the nodes on your cluster are CUDA-capable).

这篇关于任何优势MPI + CUDA超过只是纯粹的MPI？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！