instruction for install mpich2 and cuda4.2 in ubuntu

cuda_install_instruction.sh

sudo -i
apt-get update
apt-get -y install binutils gcc make linux-source linux-headers-$(uname -r) mpich2 nfs-server nfs-client
cd /usr/local/src/
wget http://developer.download.nvidia.com/compute/cuda/4_2/rel/toolkit/cudatoolkit_4.2.9_linux_64_ubuntu11.04.run
wget http://developer.download.nvidia.com/compute/cuda/4_2/rel/drivers/devdriver_4.2_linux_64_295.41.run
wget http://developer.download.nvidia.com/compute/cuda/4_2/rel/sdk/gpucomputingsdk_4.2.9_linux.run
chmod 755 /usr/local/src/*.run
cd /usr/local/src/
./cudatoolkit_4.2.9_linux_64_ubuntu11.04.run
echo "/usr/local/cuda/lib64" >> /etc/ld.so.conf
echo "/usr/local/cuda/lib" >> /etc/ld.so.conf
ldconfig
./devdriver_4.2_linux_64_295.41.run --kernel-source-path=/usr/src/linux-headers-$(uname -r)
echo "PATH=$PATH:/usr/local/cuda/bin" >> /etc/profile
echo "export PATH" >> /etc/profile
cd /usr/local/src/
./gpucomputingsdk_4.2.9_linux.run
cp ~/NVIDIA_GPU_Computing_SDK/C/lib/* /usr/local/lib/
cp ~/NVIDIA_GPU_Computing_SDK/C/common/inc/cutil*.h /usr/local/include/
cp ~/NVIDIA_GPU_Computing_SDK/C/common/inc/paramgl.h /usr/local/include/
cp ~/NVIDIA_GPU_Computing_SDK/C/common/inc/rendercheck_*.h /usr/local/include/

NFS server 設定 只要集中式的 server 設定即可

vi /etc/exports

按下 i 插入一行資料 , 接著按下ESC 再輸入 :wq! 存檔

/tmp *(rw,sync,no_subtree_check)

重開 NFS server

/etc/init.d/nfs-kernel-server restart

server 上面的 host file 設定

vi /tmp/hostfile

設定 client 的 private dns (格式會很像 ip-10-16-19-179.ec2.internal)

ip-10-16-19-179.ec2.internal

client 的設定

mount -t nfs 集中式的serverIP:/tmp /tmp