为 dask_jobqueue 创建 local_directory

Create local_directory for dask_jobqueue

我正在尝试 运行 在使用 NFS 进行存储的 HPC 系统上进行 dask。因此,我想配置 dask 以将本地存储用于临时 space。每个集群节点都有一个所有用户都可以写入的 /scratch/ 文件夹,其中包含将暂存文件放入 /scratch/<username>/<jobid>/.

的说明

我有一些代码是这样配置的:

import dask_jobqueue
from distributed import Client

cluster = dask_jobqueue.SLURMCluster(
            queue = 'high',
            cores = 24,
            memory = '60GB',
            walltime = '10:00:00',
            local_directory = '/scratch/<username>/<jobid>/'
)

cluster.scale(1)
client = Client(cluster)

但是,我有一个问题。该目录不存在(既因为我不知道客户端将在哪个节点上,又因为它是根据 SLURM 作业 ID 创建的,它始终是唯一的)所以我的代码失败了:

Process Dask Worker process (from Nanny):
Traceback (most recent call last):
  File "/home/lsterzin/anaconda3/lib/python3.7/multiprocessing/process.py", line 297, in _bootstrap
    self.run()
  File "/home/lsterzin/anaconda3/lib/python3.7/multiprocessing/process.py", line 99, in run
    self._target(*self._args, **self._kwargs)
  File "/home/lsterzin/anaconda3/lib/python3.7/site-packages/distributed/process.py", line 191, in _run
    target(*args, **kwargs)
  File "/home/lsterzin/anaconda3/lib/python3.7/site-packages/distributed/nanny.py", line 699, in _run
    worker = Worker(**worker_kwargs)
  File "/home/lsterzin/anaconda3/lib/python3.7/site-packages/distributed/worker.py", line 497, in __init__
    self._workspace = WorkSpace(os.path.abspath(local_directory))
  File "/home/lsterzin/anaconda3/lib/python3.7/site-packages/distributed/diskutils.py", line 118, in __init__
    self._init_workspace()
  File "/home/lsterzin/anaconda3/lib/python3.7/site-packages/distributed/diskutils.py", line 124, in _init_workspace
    os.mkdir(self.base_dir)
FileNotFoundError: [Errno 2] No such file or directory: '/scratch/<user>/<jobid>'

我无法在不知道 dask worker 运行 将在哪个节点的情况下创建目录,并且在目录已经存在的情况下我无法使用 dask_jobqueue 创建集群。解决此问题的最佳方法是什么?

我认为这可以用 /scratch/$USER/$SLURM_JOB_ID 来完成。如果这不起作用,也许可以通过配置文件定义 local-directoryhttps://jobqueue.dask.org/en/latest/configuration-setup.html#local-storage

示例配置也可能对您有用: https://jobqueue.dask.org/en/latest/configurations.html

感谢@lsterzinger 提出的措辞恰当的问题

我在这里推送了一个可能有帮助的修复:https://github.com/dask/distributed/pull/3928

我们看看社区怎么说