问题描述
如果我将单个字节从 CUDA 内核传输到 PCI-E 到主机(零拷贝内存)，与传输 200 兆字节的数据相比，它的速度要慢多少?
If I transfer a single byte from a CUDA kernel to PCI-E to the host (zero-copy memory), how much is it slow compared to transferring something like 200 Megabytes?
我想知道，因为我知道通过 PCI-E 传输对于 CUDA 内核来说很慢，所以我想知道的是:如果我只传输一个字节或大量数据，它会改变什么吗?或者可能由于内存传输是批量"执行的，传输单个字节相对于传输 200 MB 而言非常昂贵且无用?
What I would like to know, since I know that transferring over PCI-E is slow for a CUDA kernel, is: does it change anything if I transfer just a single byte or a huge amount of data? Or perhaps since memory transfers are performed in "bulks", transferring a single byte is extremely expensive and useless with respect to transferring 200 MBs?
推荐答案
希望这张图片能解释一切.数据由 CUDA 样本中的 bandwidthTest 生成.硬件环境为 PCI-E v2.0、Tesla M2090 和 2x Xeon E5-2609.请注意，两个轴都是对数刻度.
Hope this pic explain everything. The data is generated by bandwidthTest in CUDA samples. The hardware environment is PCI-E v2.0, Tesla M2090 and 2x Xeon E5-2609. Please note both axises are in log scale.
鉴于此图，我们可以看到启动传输请求的开销需要一个恒定的时间.对数据的回归分析得出 H2D 的估计开销时间为 4.9us，D2H 为 3.3us，D2D 为 3.0us.
Given this figure, we can see that the overhead of launching a transfer request takes a constant time. Regression analysis on the data gives an estimated overhead time of 4.9us for H2D, 3.3us for D2H and 3.0us for D2D.
这篇关于CUDA - 通过 PCI-E 传输的速度有多慢?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！