목록CUDA (4)

HPE CRAY 자료 공유

[Ubuntu] Cray Lustre client build

아래의 환경에서 Cray Lustre Client 빌드 과정을 소개 한다. - OS: Ubuntu 20.04.6 LTS- Lustre Client: Cray Lustre 2.15.B15  1. 의존성 패키지 설치# apt update# apt install module-assistant libreadline-dev debhelper \  dpatch automake pkg-config libtool libyaml-dev libnl-genl-3-dev \  libselinux-dev libsnmp-dev bzip2 quilt \  linux-headers-$(uname -r) rsync libssl-dev libpython3-dev swig \  tk tk-dev bison tcl tcl-dev  grap..
STOGAGES/LUSTRE 2024. 8. 28. 20:03
[GPU] Cuda Samples Utilities

- 소스코드 다운 $ wget -O cuda-samples-12.4.1.tar.gz https://github.com/NVIDIA/cuda-samples/archive/refs/tags/v12.4.1.tar.gz$ tar xvzf cuda-samples-12.4.1.tar.gz $ cd cuda-samples-12.4.1 $ cd Samples/1_Utilities/bandwidthTest - 빌드 방법:$ CUDA_PATH=/usr/local/cuda-11.8 make TARGET_ARCH=x86_64  1. bandwidthTest# ./bandwidthTest  [CUDA Bandwidth Test] - Starting... Running on...  Device 0: NVIDIA A100-SXM4..
SYSTEMS/GPU 2024. 5. 2. 15:56
[slurm] rpmbuild with gpu

※ slurm gres.conf 사용을 위한 간단한 예제 - cuda toolkit 설치 $ wget https://developer.download.nvidia.com/compute/cuda/11.8.0/local_installers/cuda_11.8.0_520.61.05_linux.run $ sudo sh cuda_11.8.0_520.61.05_linux.run - rpmbuild 옵션에 "--with-nvml"을 추가 $ rpmbuild --define "_with_nvml --with-nvml=/usr/local/cuda-11.8" -ta slurm-22.05.6.tar.bz2 - gpu 라이브러리 포함 확인 $ cd ${HOME}/rpmbuild/RPMS/x86_64 $ rpm -qlp slur..
Applications/Scheduler 2023. 11. 14. 15:25
[Ubuntu] Slurm source install

1. user 생성 # export MUNGEUSER=966 # groupadd -g $MUNGEUSER munge # useradd -m -c "MUNGE Uid 'N' Gid Emporium" -d /var/lib/munge -u $MUNGEUSER -g munge -s /sbin/nologin munge # export SLURMUSER=967 # groupadd -g $SLURMUSER slurm # useradd -m -c "SLURM workload manager" -d /var/lib/slurm -u $SLURMUSER -g slurm -s /bin/bash slurm 2. Dependency Packages 설치 # apt install -y munge libmunge-dev libmung..
Applications/Scheduler 2021. 9. 14. 19:16
이전 Prev 1 Next 다음