dask-iclx 0.5.1

No project description provided

dask-iclx

Adapts the dask-lxplus package to enable jobs to run on the IC HTCondor cluster.

Setup

pip install dask-iclx

Usage

from distributed import Client
from dask_iclx import ICCluster
import socket

with ICCluster(
    cores = 1,
    memory = '3000MB',
    disk = '10GB',
    death_timeout = '60',
    lcg = False,
    nanny = False,
    container_runtime = 'none',
    log_directory = '/vols/experiment/username/dask-logs',
    scheduler_options = {
        'port': 60000,
        'host': socket.gethostname(),
    },
    job_extra = {
        "+MaxRuntime": "1200",
    },
    name="ClusterName",
) as cluster:
    n_workers = 1
    with Client(cluster) as client:
        futures = []
        cluster.scale(n_workers)
        for _ in range(n_workers):
            f = client.submit(lambda: socket.gethostname())
            futures.append(f)
        print(client.gather(futures))  # ['lxb10.hep.ph.ic.ac.uk']

IC/CERN extras

There are a few changes in the wrapper to address some of the particular features of the IC HTCondor cluster, but there are also a few changes to detail here.

Options

• lcg: If set to True this will validate and use the LCG python environment per the managed LCG releases. An example use would be to do the following before running:

  source /cvmfs/sft.cern.ch/lcg/views/LCG_107/x86_64-el9-gcc14-opt/setup.sh
  

• container_runtime: Can be set to "singularity" or "none". If a runtime is needed for the worker, this selects which will be used for the HTCondor job the worker runs. In principle it should not be necessary when using lcg and should therefore be set to "none". Default though is "singularity".

• worker_image: The image that will be used if container_runtime is defined to use one. The default defined in jobqueue-ic.yaml.

• name: Optionally set a string that will identify the jobs in HTCondor.