-
Notifications
You must be signed in to change notification settings - Fork 364
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
update Summit machine config #6336
Conversation
|
<environment_variables compiler="gnu.*" mpilib="spectrum-mpi"> | ||
<env name="ADIOS2_ROOT">$SHELL{if [ -z "$ADIOS2_ROOT" ]; then echo /gpfs/alpine/cli115/world-shared/3rdparty/adios2/2.9.1/spectrum-mpi-10.4.0.3/gcc-9.1.0; else echo "$ADIOS2_ROOT"; fi}</env> | ||
</environment_variables> | ||
<!-- <environment_variables compiler="ibm.*" mpilib="spectrum-mpi"> --> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is it worth keeping this commented-out code?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I went back and fort about this, but I think it makes sense to just remove it. Just pushed a commit for this.
</environment_variables> | ||
<environment_variables compiler="gnu.*" mpilib="spectrum-mpi"> | ||
<env name="ADIOS2_ROOT">$SHELL{if [ -z "$ADIOS2_ROOT" ]; then echo /gpfs/alpine/cli115/world-shared/3rdparty/adios2/2.9.1/spectrum-mpi-10.4.0.3/gcc-9.1.0; else echo "$ADIOS2_ROOT"; fi}</env> | ||
</environment_variables> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is the above block a leftover from previous clean-up? I remember I had to manually remove the similar block of script for chrysalis to build. Asking in case a replacement is needed for this change.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Oh, reviewing this before reading the PR msg. All sounds good.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Replacement is not needed for what I plan to do on Summit. If the need for Adios arises we can buidl a new local version and add env variables here again.
@whannah1, I noticed that a hard-coded path is used for "DIN_LOC_ROOT" and similar entries in "config_machines.xml". This might cause failures for users not in the "atm146" group. I wonder if this is intentional. |
If you look at the old path you'll notice that it was also hardcoded to a specific project "cli115", but this didn't cause problems because everyone using summit for E3SM/SCREAM was on that project. I'm not aware of a better place to put the inputdata, but given that it's in the "world-shared" path I think other people can access it. Also, I'm happy to add other people to the atm146 project to run simulations on Summit. This allocation is associated with the SummitPLUS proposal I submitted for doing development and long simulations with the new MMF configuration, but since I'm already behind on burning this allocation I think it's fine for others to use it for related E3SM dev or sims. |
@whannah1, you're right that it is accessible to users outside of the "atm146" group. One thing you might need to do is add write permission for non-atm146 users under "/gpfs/alpine2/atm146/world-shared/e3sm". I ran several E3SM tests but they failed due to permission issues. |
Other than the file permission issue, everything looks good to me. I think I will merge this PR soon if there are no further concerns. |
That's good to know - I'll figure out how to fix this. |
@grnydawn you forgot to merge this to next. That has to be done at the terminal. |
Update the Summit machine configuration to support the 2024 SummitPLUS atm146 allocation. Verified to work with GNU compoiler, which is the only one planned to be used.
NOTE: the local adios library will need to be rebuilt, but it is currently not needed.