Difference between revisions of "BwUniCluster2.0/Software/Python Dask"

From bwHPC Wiki
Jump to: navigation, search
Line 5: Line 5:
   
 
== Installation ==
 
== Installation ==
Use on of our pre-configured Python modules and load them with 'module load ...'. You have to install the packages 'dask' and 'das-jobqueue' if your are yousing an own conda environment.
+
Use on of our pre-configured Python modules and load them with 'module load ...'. You have to install the packages 'dask' and 'das-jobqueue' if your are yousing an own conda environment.
  +
  +
== Using Dask ==
  +
In Python, execute the following commands:
  +
  +
<pre>
  +
>>> from dask_jobqueue import SLURMCluster
  +
>>> cluster = SLURMCluster(cores=XX, memory='XX GB', queue='XXX')
  +
</pre>
 
----
 
----
 
[[Category:bwUniCluster_2.0|Access]][[Category:Access|bwUniCluster 2.0]]
 
[[Category:bwUniCluster_2.0|Access]][[Category:Access|bwUniCluster 2.0]]

Revision as of 10:45, 19 April 2020

This guide explains how to use Python Dask and dask-jobqueue on bwUniCluster2.0.

1 Installation

Use on of our pre-configured Python modules and load them with 'module load ...'. You have to install the packages 'dask' and 'das-jobqueue' if your are yousing an own conda environment.

2 Using Dask

In Python, execute the following commands:

>>> from dask_jobqueue import SLURMCluster
>>> cluster = SLURMCluster(cores=XX, memory='XX GB', queue='XXX')