Running jobs: Difference between revisions

Jump to navigation Jump to search
Remove architecture-specific recommendations for Cedar; see the Cedar page instead
(change URL to link directly to dependency option)
(Remove architecture-specific recommendations for Cedar; see the Cedar page instead)
Line 490: Line 490:
<tab name="Cedar">
<tab name="Cedar">
Jobs may not be submitted from directories on the /home filesystem on Cedar; the maximum duration for a job is 28 days. This is to reduce the load on that filesystem and improve the responsiveness for interactive work. If the command <code>readlink -f $(pwd) | cut -d/ -f2</code> returns <code>home</code>, you are not permitted to submit jobs from that directory. Transfer the files from that directory either to a /project or /scratch directory and submit the job from there.
Jobs may not be submitted from directories on the /home filesystem on Cedar; the maximum duration for a job is 28 days. This is to reduce the load on that filesystem and improve the responsiveness for interactive work. If the command <code>readlink -f $(pwd) | cut -d/ -f2</code> returns <code>home</code>, you are not permitted to submit jobs from that directory. Transfer the files from that directory either to a /project or /scratch directory and submit the job from there.
<!--T:164-->
Cedar has two distinct CPU architectures available: [https://en.wikipedia.org/wiki/Broadwell_(microarchitecture) Broadwell] and [https://en.wikipedia.org/wiki/Skylake_(microarchitecture) Skylake]. Users requiring a specific architecture can request it when submitting a job using the <code>--constraint</code> flag. Note that the names should be written all in lower-case, <code>skylake</code> or <code>broadwell</code>.
<!--T:157-->
An example job requesting the <code>skylake</code> feature on Cedar:
<pre>
#!/bin/bash
#SBATCH --account=def-someuser
#SBATCH --constraint=skylake
#SBATCH --time=0:5:0
# Display CPU-specific information with 'lscpu'.
# Skylake CPUs will have 'avx512f' in the 'Flags' section of the output.
lscpu
</pre>
<!--T:158-->
Keep in mind that a job which would have obtained an entire node for itself by specifying for example <code>#SBATCH --cpus-per-task=32</code> will now share the remaining 16 CPU cores with another job if it happens to use a Skylake node; if you wish to reserve the entire node you will need to request all 48 cores or add the <code>#SBATCH --constraint=broadwell</code> option to your job script.
<!--T:159-->
<i>If you are unsure if your job requires a specific architecture, do not use this option.</i> Jobs that do not specify a CPU architecture can be scheduled on either Broadwell or Skylake nodes, and will therefore generally start earlier.
</tab>
</tab>


cc_staff
127

edits

Navigation menu