MPI:阻止与非阻止 [英] MPI: blocking vs non-blocking

问题描述

我很难理解MPI中阻塞通信和非阻塞通信的概念.两者之间有什么区别?优点和缺点是什么?

I am having trouble understanding the concept of blocking communication and non-blocking communication in MPI. What are the differences between the two? What are the advantages and disadvantages?

推荐答案

使用 MPI_Recv() .在通信完成之前，这些函数不会返回(即，它们阻塞).稍微简化一下，这意味着传递给MPI_Send()的缓冲区可以重复使用，这是因为MPI将其保存在某个位置，或者是因为它已被目标接收.同样，MPI_Recv()将在接收缓冲区中已填充有效数据时返回.

Blocking communication is done using MPI_Send() and MPI_Recv(). These functions do not return (i.e., they block) until the communication is finished. Simplifying somewhat, this means that the buffer passed to MPI_Send() can be reused, either because MPI saved it somewhere, or because it has been received by the destination. Similarly, MPI_Recv() returns when the receive buffer has been filled with valid data.

相反，使用 和 MPI_Irecv() .即使通信尚未完成，这些函数也会立即返回(即，它们不会阻塞).您必须致电 MPI_Wait() 或 MPI_Test() 以查看通信是否已完成.

In contrast, non-blocking communication is done using MPI_Isend() and MPI_Irecv(). These function return immediately (i.e., they do not block) even if the communication is not finished yet. You must call MPI_Wait() or MPI_Test() to see whether the communication has finished.

在足够的情况下使用阻塞通信，因为它比较容易使用.必要时使用无阻塞通信，例如，您可以调用MPI_Isend()，进行一些计算，然后再进行MPI_Wait().这样可以使计算和通信重叠，从而通常可以提高性能.

Blocking communication is used when it is sufficient, since it is somewhat easier to use. Non-blocking communication is used when necessary, for example, you may call MPI_Isend(), do some computations, then do MPI_Wait(). This allows computations and communication to overlap, which generally leads to improved performance.

请注意，集体通信(例如全减少)仅在其阻塞版本(最多MPIv2)中可用. IIRC，MPIv3引入了非阻塞的集体通信.

Note that collective communication (e.g., all-reduce) is only available in its blocking version up to MPIv2. IIRC, MPIv3 introduces non-blocking collective communication.

此处.

这篇关于MPI:阻止与非阻止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！