最合适的MPI_Datatype的＆QUOT;块分解＆QUOT;？ [英] Most appropriate MPI_Datatype for "block decomposition"?
问题描述
从乔纳森Dursi 的帮助和的<a href =http://stackoverflow.com/users / 196561 / osgx> osgx ，我现在已经做了的处理中排分解：
With the help from Jonathan Dursi and osgx, I've now done the "row decomposition" among the processes:
现在，我想试试块分解的方式（如下图所示）：
Now, I'd like to try the "block decomposition" approach (pictured below):
怎么做呢？这一次， MPI_Datatype 将是必要的，对不对？ 哪些数据类型是最合适的/容易使用？或可以将它振振有词没有一个数据类型来完成？
How should one go about it? This time, the MPI_Datatype will be necessary, right? Which datatype would be most appropriate/easy to use? Or can it plausibly be done without a datatype?
推荐答案
您总是可以凑合未经只创建一个缓冲区，复制缓冲区的基本类型的计数数据类型;这是概念上最简单的。另一方面，它慢，它实际上涉及code的更多的行。尽管如此，它可以方便的时候你要得到的东西的工作，然后你就可以实现沿边的数据类型-Y版本，并确保你得到同样的答案。
You can always make do without a datatype by just creating a buffer and copying the buffer as count of the underlying type; that's conceptually the simplest. On the other hand, it slower and it actually involves a lot more lines of code. Still, it can be handy when you're trying to get something to work, and then you can implement the datatype-y version along side that and make sure you're getting the same answers.
有关鬼细胞充盈，在我方向你不需要一个类型，因为它是类似于你一直在做什么;但你可以使用一个， MPI_Type_contiguous ，刚刚指定某种类型的计数（你可以在你的发送/ recv的反正这样做）。
For the ghost-cell filling, in the i direction you don't need a type, as it's similar to what you had been doing; but you can use one, MPI_Type_contiguous, which just specifies a count of some type (which you can do anyway in your send/recv).
有关在J方向鬼细胞填充，可能最简单的就是使用的 MPI_Type_Vector 。如果你发送的最右边的列（说）其中i = 0..N-1，J = 0..M-1要发送一个向量数量= N，块大小= 1，步幅= M阵列。也就是说，你发送1值的数块，分别由M值阵列中分离出来。
For ghost-cell filling in the j direction, probably easiest is to use MPI_Type_Vector. If you're sending the rightmost column of (say) an array with i=0..N-1, j=0..M-1 you want to send a vector with count=N, blocksize=1, stride=M. That is, you're sending count chunks of 1 value, each separated by M values in the array.
您也可以使用 MPI_Type_create_subarray 拔出要在阵列刚刚区域;这可能是在这种情况下，有点矫枉过正。
You can also use MPI_Type_create_subarray to pull out just the region of the array you want; that's probably a little overkill in this case.
现在，如果在你的previous问题，你希望能够在某些时候所有的子阵列收集到一个处理器，你可能会使用子阵，问题的一部分，在这里回答： <一href=\"http://stackoverflow.com/questions/5585630/mpi-type-create-subarray-and-mpi-gather\">MPI_Type_create_subarray和MPI_Gather 的。请注意，如果你的阵列块的大小不同，不过，事情开始变得有点tricker。
Now, if as in your previous question you want to be able at some point to gather all the sub-arrays onto one processor, you'll probably be using subarrays, and part of the question is answered here: MPI_Type_create_subarray and MPI_Gather . Note that if your array chunks are of different sizes, though, then things start getting a little tricker.
（事实上，你为什么做收集到一个处理器呢？这最终会成为一个可扩展的瓶颈。如果你这样做是对I / O，一旦你熟悉的数据类型，你可以使用MPI-IO这个..）
(Actually, why are you doing the gather onto one processor, anyway? That'll eventually be a scalability bottleneck. If you're doing it for I/O, once you're comfortable with data types, you can use MPI-IO for this..)
这篇关于最合适的MPI_Datatype的＆QUOT;块分解＆QUOT;？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！
