MPI约定用于行和列的索引 [英] MPI convention for index of rows and columns

问题描述

我正在使用MPI解决PDE.为此，我将2D域分解为不同的单元格(每个单元格的大小分别为xcell,ycell和xcell = size_x_domain/(number of X subdomains)和ycell = size_y_domain/(number of Y subdomains).

因此，我正在使用number of processes = (number of X subdomains)*(number of Y subdomains)

运行代码

相对于顺序版本而言，收益是我在代表子域的每个进程之间进行通信.

下图显示了我的分解过程，其中包括8个进程(X包含2个子域，Y包含4个子域):

(xs,xe)代表单元格的x_start和x_end， (ys,ye)代表单元格的y_start和y_end

我想知道是否必须将 i索引设置为行索引并将 j设置为列索引吗?

将第一个索引放置在行中，将第二个索引放置在列中是一般规则吗? (例如，使用C，Fortran和Matlab语言或更多语言)

感谢您的帮助.

解决方案

我不确定，但也许尝试对所有4种通信使用不同的标志.

I am using MPI for solving PDE. For this, I breakdown the 2D domain into different cells (size of each of these cells is "xcell,ycell" with xcell = size_x_domain/(number of X subdomains) and ycell = size_y_domain/(number of Y subdomains).

So, I am running the code with number of processes = (number of X subdomains)*(number of Y subdomains)

The gain relatively to sequential version is that I communicate between each process representing the sub-domains.

Here a figure illustrating my breakdown with 8 processes (2 subdomains for X and 4 for Y) :

(xs,xe) represent x_start and x_end of the cell, (ys,ye) represent y_start and y_end of the cell

I would like to know if I have to set, into x(i,j) array, i the index as row index and j as column index ?

Is it a general rule to put the first index for row and the second one for column ? ( for example, in C, Fortran and Matlab language or maybe more)

Thanks for your help.

解决方案

I'm not sure, but maybe try having a different flag for all 4 of the communications.

这篇关于MPI约定用于行和列的索引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！