阿里ECS GPU机型如何安装驱动(系统:CentOS7.3 GPU: Nvidia P100)

  

一、配置DNS以及百胜
1,配置DNS

  
 <代码> root@gpu-test-01 ~ # vim/etc/resolv.conf
  命名服务器223.5.5.5
  命名服务器114.114.114.114
  尝试选择超时:2:3 single-request-reopen旋转
  说明:我这里配置了两个外部DNS 223.5.5.5以及114.114.114.114
  (root@gpu-test-01 ~) # chattr +我/etc/reslov.conf  
  

2,配置百胜

  
 <代码> root@gpu-test-01 ~ # cd/etc/yum.repos.d/[root@gpu-test-01 yum.repos。d] # rm射频/*
  [root@gpu-test-01 yum.repos。d] # mv/etc/yum.repos.d/*/tmp
  [root@gpu-test-01 yum.repos。d] # wget - o/etc/yum.repos.d/CentOS-Base。回购http://mirrors.aliyun.com/repo/centos - 7.回购
  [root@gpu-test-01 yum.repos。d] # wget - o/etc/yum.repos.d/epel。回购http://mirrors.aliyun.com/repo/epel - 7.回购
  [root@gpu-test-01 yum.repos。d] https://us.download.nvidia.cn/tesla/418.67/nvidia # wget -诊断接头司机-地方-回购rhel7 - 418.67 - 1.0 - 1. - x86_64.rpm
  [root@gpu-test-01 yum.repos。d] # yum安装nvidia -诊断接头司机-地方-回购rhel7 - 418.67 - 1.0 - 1. - x86_64。rpm - y
  [root@gpu-test-01 yum.repos。d] # mv nvidia -诊断接头-司机——当地回购rhel7 - 418.67 - 1.0 - 1. - x86_64。rpm/tmp/ 
  

二,下载驱动包

  

1,下载P100/P4驱动:

  
 <代码> [root@gpu-test-01 ~] # wget http://us.download.nvidia.com/tesla/396.44/NVIDIA-Linux-x86_64-396.44.run  
  

2,下载内核开发包:

  
 <代码> [root@gpu-test-01 ~] # wget https://buildlogs.centos.org/c7.1611.u/kernel/20170620132051/3.10.0-514.21.2.el7.x86_64/kernel-devel-3.10.0-514.21.2.el7.x86_64.rpm  
  

3,下载cuda包:(如果使用yum来装cuda-drivers,这一步也可以忽略)

  
 <代码> [root@gpu-test-01 ~] # wget https://developer.nvidia.com/compute/cuda/9.1/Prod/local_installers/cuda_9.1.85_387.26_linux  
  

三,配置信息

  

下载并安装内核对应版本的kernel-devel和kernel-header包

  
 <代码> [root@gpu-test-01 ~] # rpm -ivh kernel-devel-3.10.0-514.21.2.el7.x86_64.rpm
  准备……# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # (100%)
  更新/安装…
  1:kernel-devel-3.10.0-514.21.2。el7 # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # (100%)
  (root@gpu-test-01 ~) # sudo rpm qa | grep (uname - r)美元
  kernel-headers-3.10.0-514.21.2.el7.x86_64
  kernel-3.10.0-514.21.2.el7.x86_64
  kernel-devel-3.10.0-514.21.2.el7.x86_64  
  

说明:kernel-devel和内核版本不一致会导致在安装驱动转速过程中司机编译出错。您可以在实例里运行rpm qa | grep内核检测版本是否一致。确认版本后,再重新安装驱动。

  
 <代码> root@gpu-test-01 ~ # sh NVIDIA-Linux-x86_64-396.44.run
  按照引导一路下一步: 
  

验证下是否安装成功:

  
 <代码> root@gpu-test-01 ~ # nvidia-smi
  2019年6月22日18:39:14坐下
  +-----------------------------------------------------------------------------+
  | | NVIDIA-SMI 396.44驱动程序版本:396.44
  |-------------------------------+----------------------+----------------------+
  | GPU名字Persistence-M | Bus-Id Disp.A | Uncorr波动。ECC |
  |风扇温度性能压水式反应堆:使用/帽| |的内存GPU-Util计算m . |
  |===============================+======================+======================|
  | 0特斯拉P100-PCIE……了| | 0 | 00000000:00:08.0
  | N/A 33 c P0 27 16280 w/250 w | 0 mib/mib默认| | 4%
  +-------------------------------+----------------------+----------------------+
  
  +-----------------------------------------------------------------------------+
  | |过程:GPU内存
  | | GPU PID型进程名称用法
  |=============================================================================|
  发现| |没有运行流程
  + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - +  
  

到此驱动已经安装完成。

阿里ECS GPU机型如何安装驱动(系统:CentOS7.3 GPU: Nvidia P100)