BwUniCluster2.0/FAQ - broadwell partition: Difference between revisions

From bwHPC Wiki
Jump to navigation Jump to search
m (R Barthel moved page BwUniCluster broadwell partition to FAQ - bwUniCluster broadwell partition: this is a FAQ therefore renaming)
No edit summary
Line 9: Line 9:
== How to login to broadwell login nodes? ==
== How to login to broadwell login nodes? ==
* You can directly login on broadwell partition login nodes using
* You can directly login on broadwell partition login nodes using
<pre>
$ username@uc1e.scc.kit.edu
$ ssh username@uc1e.scc.kit.edu
</pre>
* If you login to the ''old'' uc1 login nodes, even though you can use broadwell nodes using the same procedure as 'compute nodes'.
* If you login to the ''old'' uc1 login nodes, even though you can use broadwell nodes using the same procedure as 'compute nodes'.


Line 17: Line 19:
On uc1 (old) login nodes:
On uc1 (old) login nodes:
<pre>
<pre>
icc/ifort -axCORE-AVX2
$ icc/ifort -axCORE-AVX2 [-further_options]
</pre>
</pre>
On uc1e (extension) login nodes:
On uc1e (extension) login nodes:
<pre>
<pre>
icc/ifort -xHost
$ icc/ifort -xHost [-further_options]
</pre>
</pre>


Line 27: Line 29:
On uc1e (= extension) login nodes:
On uc1e (= extension) login nodes:
<pre>
<pre>
icc/ifort -xAVX -axCORE-AVX2
$ icc/ifort -xAVX -axCORE-AVX2 [-further_options]
</pre>
</pre>


Line 39: Line 41:
The submitted job will be distributed either way to the broadwell nodes if specified correctly, i.e.:
The submitted job will be distributed either way to the broadwell nodes if specified correctly, i.e.:
<pre>
<pre>
msub -q multinode
$ msub -q multinode
</pre>
</pre>



Revision as of 11:22, 18 September 2017

FAQs concerning best practice of bwUniCluster broadwell partition (aka "extension" partition).

Login

Are there separate login nodes for the bwUniCluster broadwell partition?

  • Yes, but primarily to be used for compiling code.

How to login to broadwell login nodes?

  • You can directly login on broadwell partition login nodes using
$ ssh username@uc1e.scc.kit.edu
  • If you login to the old uc1 login nodes, even though you can use broadwell nodes using the same procedure as 'compute nodes'.


Compilation

How to compile code on broadwell (extension) nodes?

On uc1 (old) login nodes:

$ icc/ifort -axCORE-AVX2 [-further_options]

On uc1e (extension) login nodes:

$ icc/ifort -xHost [-further_options]

How to compile the same code on old and extension partition?

On uc1e (= extension) login nodes:

$ icc/ifort -xAVX -axCORE-AVX2 [-further_options]

What happens with code compiled for old partition whic is running on the extension partition?

Code will run but significantly slower since AVX2 will not be used. Please recompile your code accordingly.


Job execution

How to submit jobs to the broadwell (= extension) partition

The submitted job will be distributed either way to the broadwell nodes if specified correctly, i.e.:

$ msub -q multinode