aistabci / abci-docs Goto Github PK

• /ja/docs/01.md#software
• /en/docs/01.md#software

Singularity Endpoint のページの日本語版に適用された差分を英語版にも適用します。

#235

• /ja/mkdocs.yaml
• /ja/docs/index.md
• /en/mkdocs.yaml
• /en/docs/index.md

/en/docs/02.md などでは "Usage Manager"、https://portal.abci.ai/docs/portal/en/03/ では "users who are granted the user administrator authority" という表現になっています。そもそも "Usage Manager", "user administrator" という表現がおかしいので、"Group Administrator (users who are granted the group administrator authority)" などという表現で統一してください。

Hi, thanks for your nice work.

Now I am a user of ABCI server. I found there is only 200G disk space in home directory for each user. In order to run a large model on ABCI server, I have applied 5T group disk space. But I cannot find any guide about how to use Group disk in ABCI document.

Could you give me some advices about how to mount or use group disk?

2019-11-26現在、module avail によるとgccに関しては gcc/4.8.5 , gcc/7.3.0, gcc/7.4.0 が準備されているようですが、Tips[1]にはそれが反映されていないように見えます。

[1] https://docs.abci.ai/ja/tips/gcc-7.3.0/

• #216

• #207

• /ja/docs/07.md
• /en/docs/07.md

• /ja/docs/system-updates.md
• /en/docs/system-updates.md

• appendix/ssh-access.md を追加して記述。
• 機能を説明することとし、torch.distributed の例などは tips 以下に追加。後者は milestone に含めない。

1.バッチジョブの投入
[username@es1 ~]$ qsub -g grpname -l rt_F=2 run.sh
Your job 1000000 ("run.sh") has been submitted

2.ジョブに割り当てられた計算ノードの確認
[username@es1 ~]$ qstat -j 1000000
(snip)
exec_host_list 1: g0001:80, g0002:80
(snip)

3.環境変数の設定
[username@es1 ~]$ export JOB_ID=1000000
[username@es1 ~]$ export SGE_TASK_ID=undefined

4.計算ノードでnvidia-smiの実行
[username@es1 ~]$ qrsh -inherit g0001 nvidia-smi
Wed Oct 21 16:01:12 2020
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.33.01 Driver Version: 440.33.01 CUDA Version: 10.2 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 Tesla V100-SXM2... On | 00000000:3D:00.0 Off | 0 |
| N/A 31C P0 38W / 300W | 0MiB / 16160MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 1 Tesla V100-SXM2... On | 00000000:3E:00.0 Off | 0 |
| N/A 29C P0 42W / 300W | 0MiB / 16160MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 2 Tesla V100-SXM2... On | 00000000:B1:00.0 Off | 0 |
| N/A 30C P0 42W / 300W | 0MiB / 16160MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 3 Tesla V100-SXM2... On | 00000000:B2:00.0 Off | 0 |
| N/A 32C P0 42W / 300W | 0MiB / 16160MiB | 0% Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| No running processes found |
+-----------------------------------------------------------------------------+
[username@es1 ~]$

• #209

I loaded modules
module load python/3.6 cuda/9.2
and then created venv, then installed cupy-cuda92, but failed to import cupy.

$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2018 NVIDIA Corporation
Built on Wed_Apr_11_23:16:29_CDT_2018
Cuda compilation tools, release 9.2, V9.2.88

$ pip3 freeze
absl-py==0.7.0
astor==0.7.1
chainer==5.2.0
cupy-cuda92==5.2.0
fastrlock==0.4
filelock==3.0.10
gast==0.2.2
grpcio==1.18.0
h5py==2.9.0
imageio==2.5.0
Keras==2.2.4
Keras-Applications==1.0.7
Keras-Preprocessing==1.0.9
Markdown==3.0.1
numpy==1.16.1
Pillow==5.4.1
protobuf==3.6.1
PyYAML==3.13
scikit-learn==0.20.2
scipy==1.2.1
six==1.12.0
sklearn==0.0
tensorboard==1.12.2
tensorflow==1.12.0
termcolor==1.1.0
Werkzeug==0.14.1

$ python3
Python 3.6.5 (default, Jun  2 2018, 15:49:50) 
[GCC 4.8.5 20150623 (Red Hat 4.8.5-16)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import cupy
Traceback (most recent call last):
  File "/fs3/home/aca10485tl/dl/lib/python3.6/site-packages/cupy/__init__.py", line 11, in <module>
    from cupy import core  # NOQA
  File "/fs3/home/aca10485tl/dl/lib/python3.6/site-packages/cupy/core/__init__.py", line 1, in <module>
    from cupy.core import core  # NOQA
ImportError: libcuda.so.1: cannot open shared object file: No such file or directory

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/fs3/home/aca10485tl/dl/lib/python3.6/site-packages/cupy/__init__.py", line 32, in <module>
    six.reraise(ImportError, ImportError(msg), exc_info[2])
  File "/fs3/home/aca10485tl/dl/lib/python3.6/site-packages/six.py", line 692, in reraise
    raise value.with_traceback(tb)
  File "/fs3/home/aca10485tl/dl/lib/python3.6/site-packages/cupy/__init__.py", line 11, in <module>
    from cupy import core  # NOQA
  File "/fs3/home/aca10485tl/dl/lib/python3.6/site-packages/cupy/core/__init__.py", line 1, in <module>
    from cupy.core import core  # NOQA
ImportError: CuPy is not correctly installed.

If you are using wheel distribution (cupy-cudaXX), make sure that the version of CuPy you installed matches with the version of CUDA on your host.
Also, confirm that only one CuPy package is installed:
  $ pip freeze

If you are building CuPy from source, please check your environment, uninstall CuPy and reinstall it with:
  $ pip install cupy --no-cache-dir -vvvv

Check the Installation Guide for details:
  https://docs-cupy.chainer.org/en/latest/install.html

original error: libcuda.so.1: cannot open shared object file: No such file or directory

単純に提供中止予定であることをwarningとして追加すること。CVEへの言及、対処方法の説明は不要。