Skip to content

Commit

Permalink
チュートリアル:OCI HPCテクニカルTips、OCI HPCチュートリアル集
Browse files Browse the repository at this point in the history
カテゴリ:OCI HPCポータル
更新内容:"Slurmによるリソース管理・ジョブ管理システム構築方法"のPMIxバージョンを5から4に変更、Intel MPI Benchmarks実行方法の各コンテンツ間の記載構成変更、これらに伴う修正、及び誤記訂正・軽微な修正
  • Loading branch information
fwiw6430 committed Jun 21, 2024
1 parent 8ab0ec3 commit 68a4dac
Show file tree
Hide file tree
Showing 8 changed files with 685 additions and 733 deletions.
4 changes: 2 additions & 2 deletions _pages/hpc.md
Original file line number Diff line number Diff line change
Expand Up @@ -192,7 +192,7 @@ HPC/機械学習ワークロードを実行する際に有益なテクニカル

- **HPL**
- **STREAM**
- **Intel MPI Benchmark**
- **Intel MPI Benchmarks**
- **NCCL Tests**

各ベンチマークの実行方法は、下表の対象シェイプ部分のリンクをクリックして参照下さい。
Expand All @@ -201,7 +201,7 @@ HPC/機械学習ワークロードを実行する際に有益なテクニカル
| :---------------------: | :--------------------------------------------------------------------------------------------------------: | :-------------------------------------------------------------------------------------------------------------------------------------: |
| **HPL** | **[Link](https://www.netlib.org/benchmark/hpl/)** | **[BM.Optimized3.36](/ocitutorials/hpc/benchmark/run-hpl/)**<br>**[BM.Standard.E5.192](/ocitutorials/hpc/benchmark/run-hpl-e5/)** |
| **STREAM** | **[Link](https://www.cs.virginia.edu/stream/)** | **[BM.Optimized3.36](/ocitutorials/hpc/benchmark/run-stream/)**<br>**[BM.Standard.E5.192](/ocitutorials/hpc/benchmark/run-stream-e5/)** |
| **Intel MPI Benchmark** | **[Link](https://www.intel.com/content/www/us/en/developer/articles/technical/intel-mpi-benchmarks.html)** | **[BM.Optimized3.36](/ocitutorials/hpc/benchmark/run-imb/)** |
| **Intel MPI Benchmarks** | **[Link](https://www.intel.com/content/www/us/en/developer/articles/technical/intel-mpi-benchmarks.html)** | **[BM.Optimized3.36](/ocitutorials/hpc/benchmark/run-imb/)** |
| **NCCL Tests** | **[Link](https://github.com/NVIDIA/nccl-tests)** | **[BM.GPU.A100-v2.8/BM.GPU4.8](/ocitutorials/hpc/benchmark/run-nccltests/)** |

## 2-2. パフォーマンス関連Tips集
Expand Down
653 changes: 290 additions & 363 deletions tutorials/_hpc/benchmark/run-imb.md

Large diffs are not rendered by default.

1 change: 1 addition & 0 deletions tutorials/_hpc/spinup-cluster-network.md
Original file line number Diff line number Diff line change
Expand Up @@ -145,6 +145,7 @@ runcmd:
- mkfs.xfs -L localscratch /dev/nvme0n1p1
- mkdir -p /mnt/localdisk
- echo "LABEL=localscratch /mnt/localdisk/ xfs defaults,noatime 0 0" >> /etc/fstab
- systemctl daemon-reload
- mount /mnt/localdisk
#
# Stop firewalld
Expand Down
1 change: 1 addition & 0 deletions tutorials/_hpc/spinup-gpu-cluster-withterraform.md
Original file line number Diff line number Diff line change
Expand Up @@ -193,6 +193,7 @@ Bastionノード構築は、 **[cloud-init](/ocitutorials/hpc/#5-11-cloud-init)*
- mkfs.xfs -L localscratch /dev/nvme/lvol0
- mkdir -p /mnt/localdisk
- echo "LABEL=localscratch /mnt/localdisk/ xfs defaults,noatime 0 0" >> /etc/fstab
- systemctl daemon-reload
- mount /mnt/localdisk
#
# Stop firewalld
Expand Down
1 change: 1 addition & 0 deletions tutorials/_hpc/spinup-hpc-cluster-withterraform.md
Original file line number Diff line number Diff line change
Expand Up @@ -166,6 +166,7 @@ Bastionノード構築は、 **[cloud-init](/ocitutorials/hpc/#5-11-cloud-init)*
- mkfs.xfs -L localscratch /dev/nvme0n1p1
- mkdir -p /mnt/localdisk
- echo "LABEL=localscratch /mnt/localdisk/ xfs defaults,noatime 0 0" >> /etc/fstab
- systemctl daemon-reload
- mount /mnt/localdisk
#
# Stop firewalld
Expand Down
337 changes: 163 additions & 174 deletions tutorials/_hpc/tech-knowhow/build-openmpi.md

Large diffs are not rendered by default.

421 changes: 227 additions & 194 deletions tutorials/_hpc/tech-knowhow/setup-slurm-cluster.md

Large diffs are not rendered by default.

Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.

0 comments on commit 68a4dac

Please sign in to comment.