---
title: Performance and Hardware Configuration
---
# Performance and Hardware Configuration
To measure performance on different NVIDIA GPUs we use CaffeNet, the Caffe reference ImageNet model.
For training, each time point is 20 iterations/minibatches of 256 images for 5,120 images total. For testing, a 50,000 image validation set is classified.
**Acknowledgements**: BVLC members are very grateful to NVIDIA for providing several GPUs to conduct this research.
## NVIDIA K40
Performance is best with ECC off and boost clock enabled. While ECC makes a negligible difference in speed, disabling it frees ~1 GB of GPU memory.
Best settings with ECC off and maximum clock speed in standard Caffe:
* Training is 26.5 secs / 20 iterations (5,120 images)
* Testing is 100 secs / validation set (50,000 images)
Best settings with Caffe + [cuDNN acceleration](http://nvidia.com/cudnn):
* Training is 19.2 secs / 20 iterations (5,120 images)
* Testing is 60.7 secs / validation set (50,000 images)
Other settings:
* ECC on, max speed: training 26.7 secs / 20 iterations, test 101 secs / validation set
* ECC on, default speed: training 31 secs / 20 iterations, test 117 secs / validation set
* ECC off, default speed: training 31 secs / 20 iterations, test 118 secs / validation set
### K40 configuration tips
For maximum K40 performance, turn off ECC and boost the clock speed (at your own risk).
To turn off ECC, do
sudo nvidia-smi -i 0 --ecc-config=0 # repeat with -i x for each GPU ID
then reboot.
Set the "persistence" mode of the GPU settings by
sudo nvidia-smi -pm 1
and then set the clock speed with
sudo nvidia-smi -i 0 -ac 3004,875 # repeat with -i x for each GPU ID
but note that this configuration resets across driver reloading / rebooting. Include these commands in a boot script to initialize these settings. For a simple fix, add these commands to `/etc/rc.local` (on Ubuntu).
## NVIDIA Titan
Training: 26.26 secs / 20 iterations (5,120 images).
Testing: 100 secs / validation set (50,000 images).
cuDNN Training: 20.25 secs / 20 iterations (5,120 images).
cuDNN Testing: 66.3 secs / validation set (50,000 images).
## NVIDIA K20
Training: 36.0 secs / 20 iterations (5,120 images).
Testing: 133 secs / validation set (50,000 images).
## NVIDIA GTX 770
Training: 33.0 secs / 20 iterations (5,120 images).
Testing: 129 secs / validation set (50,000 images).
cuDNN Training: 24.3 secs / 20 iterations (5,120 images).
cuDNN Testing: 104 secs / validation set (50,000 images).
性能和硬件配置
为了衡量不同NVIDIA GPU上的性能,我们使用CaffeNet(Caffe参考ImageNet模型)。
为了进行训练,每个时间点是256张图像的20次迭代/minibatches处理,总共5120张图像。为了进行测试,分类了50,000个图像验证集。
NVIDIA K40
关闭ECC并启用加速时钟时,性能最佳。 ECC的速度差异可忽略不计,但将其禁用可释放约1 GB的GPU内存。
禁用ECC的最佳设置和标准Caffe中的最大时钟速度:
*训练时间为26.5秒/ 20次迭代(5,120张图像)
*测试时间为100秒/验证集(50,000张图片)
使用Caffe + [cuDNN加速](http://nvidia.com/cudnn)的最佳设置:
*训练时间为19.2秒/ 20次迭代(5,120张图像)
*测试时间为60.7秒/验证集(50,000张图片)
其他设定:
*开启ECC,最高速度:训练26.7秒/ 20次迭代,测试101秒/验证集
*开启ECC,默认速度:训练31秒/ 20次迭代,测试117秒/验证集
- ECC关闭,默认速度:训练31秒/ 20次迭代,测试118秒/验证集
K40配置技巧
为了获得最佳的K40性能,请关闭ECC并提高时钟速度(后果自负)。
要关闭ECC,请执行
sudo nvidia-smi -i 0 --ecc-config = 0#对每个GPU ID用-i x重复
然后重新启动。
通过以下方式设置GPU设置的“持久性”模式
sudo nvidia-smi -pm 1
然后设置闹铃速度
sudo nvidia-smi -i 0 -ac 3004,875#对每个GPU ID用-i x重复
但请注意,此配置会在驱动程序重新加载/重新引导时重置。将这些命令包含在引导脚本中以初始化这些设置。为了简单的修复,请将这些命令添加到/ etc / rc.local
(在Ubuntu上)。
NVIDIA Titan
训练:26.26秒/ 20次迭代(5,120张图像)。
测试:100秒/验证集(50,000张图像)。
cuDNN训练:20.25秒/ 20次迭代(5,120张图像)。
cuDNN测试:66.3秒/验证集(50,000张图像)。
NVIDIA K20
训练:36.0秒/ 20次迭代(5,120张图像)。
测试:133秒/验证集(50,000张图像)。
NVIDIA GTX 770
训练:33.0秒/ 20次迭代(5,120张图像)。
测试:129秒/验证集(50,000张图像)。
cuDNN训练:24.3秒/ 20次迭代(5,120张图像)。
cuDNN测试:104秒/验证集(50,000张图像)。