3.5

GjzCV

服务器的配置

一块 K40，两块 K20

Device 0: "Tesla K40c"
  CUDA Driver Version / Runtime Version          7.5 / 7.5
  CUDA Capability Major/Minor version number:    3.5
  Total amount of global memory:                 11520 MBytes (12079136768 bytes)
  (15) Multiprocessors, (192) CUDA Cores/MP:     2880 CUDA Cores
  GPU Max Clock rate:                            745 MHz (0.75 GHz)
  Memory Clock rate:                             3004 Mhz
  Memory Bus Width:                              384-bit
  L2 Cache Size:                                 1572864 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
  Maximum Layered 1D Texture Size, (num) layers  1D=(16384), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(16384, 16384), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  2048
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 2 copy engine(s)
  Run time limit on kernels:                     No
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Enabled
  Device supports Unified Addressing (UVA):      Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 132 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

Device 1: "Tesla K20m"
  CUDA Driver Version / Runtime Version          7.5 / 7.5
  CUDA Capability Major/Minor version number:    3.5
  Total amount of global memory:                 4800 MBytes (5032706048 bytes)
  (13) Multiprocessors, (192) CUDA Cores/MP:     2496 CUDA Cores
  GPU Max Clock rate:                            706 MHz (0.71 GHz)
  Memory Clock rate:                             2600 Mhz
  Memory Bus Width:                              320-bit
  L2 Cache Size:                                 1310720 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
  Maximum Layered 1D Texture Size, (num) layers  1D=(16384), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(16384, 16384), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  2048
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 2 copy engine(s)
  Run time limit on kernels:                     No
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Enabled
  Device supports Unified Addressing (UVA):      Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 3 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

安装 cuda 的命令

sudo gdebi cuda-repo-ubuntu1404_7.5-18_amd64.deb 
sudo apt-get update
sudo apt-get install cuda

在 bash.bashrc 配置

export CUDA_HOME=/usr/local/cuda-7.5 
export LD_LIBRARY_PATH=${CUDA_HOME}/lib64 
exportPATH=${CUDA_HOME}/bin:${PATH} 

测试

$ cuda-install-samples-7.5.sh  ~ 
$ cd ~/NVIDIA_CUDA-7.5_Samples 
$ cd 1_Utilities/deviceQuery 
$ make
$ ./deviceQuery

重新编译 caffe

看来之前的 caffe 用的是 cuda7.0 编译的

➜  caffe ./examples/mnist/train_lenet.sh 
./build/tools/caffe: error while loading shared libraries: libcudart.so.7.0: cannot open shared object file: No such file or directory

编译 OpenBLAS

https://github.com/xianyi/OpenBLAS/wiki/Installation-Guide

安装 cudnn

各种编译

1

cuda，atlas，无 cudnn

GPU相关

查看 GPU 使用信息

nvidia-smi
nvidia-smi --loop=1

使用多个 GPU

caffe -gpu all

测试 (MNIST)

仅 cuda 模式下，所有 GPU(K40,K20,K20): 5'52''

Sat Mar  5 13:52:29 CST 2016
Sat Mar  5 13:58:11 CST 2016

仅 cuda 模式下，K40: 4'42''

Sat Mar  5 14:01:47 CST 2016
Sat Mar  5 14:06:29 CST 2016

cuda + OpenBLAS，K40: 4'49

Sat Mar  5 14:18:02 CST 2016
Sat Mar  5 14:22:51 CST 2016

cuda + cudnn3.0, k40: 41'', 37''，平均大概 37''

#1
Sat Mar  5 15:16:35 CST 2016
Sat Mar  5 15:17:16 CST 2016
#2
Sat Mar  5 15:19:19 CST 2016
Sat Mar  5 15:19:56 CST 2016

cuda + cudnn4.0 与 上面结果差不多
cuda + cudnn4.0 + OpenBLAS 还是差不多