Skip to content
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

Refactor how spare nodes get computed. #1816

Merged
merged 2 commits into from
Aug 16, 2017
Merged

Conversation

jgfouca
Copy link
Contributor

@jgfouca jgfouca commented Aug 15, 2017

The prior implementation was a bit too simple. Always giving the user
10 percent of nodes as spare nodes was way overkill for large jobs.

The new implementation maxes out the number of spare nodes at 10.

We also add a new variable to allow the user to pick the exact number
of spare nodes that they want.

Test suite: scripts_regression_tests T_TestRunRestart
Test baseline:
Test namelist changes:
Test status: [bit for bit, roundoff, climate changing]

Fixes [CIME Github issue #]

User interface changes?: Yes, new case variables controlling spare nodes

Update gh-pages html (Y/N)?: N

Code review: @jedwards4b

The prior implementation was a bit too simple. Always giving the user
10 percent of nodes as spare nodes was way overkill for large jobs.

The new implementation maxes out the number of spare nodes at 10.

We also add a new variable to allow the user to pick the exact number
of spare nodes that they want.
@jgfouca
Copy link
Contributor Author

jgfouca commented Aug 16, 2017

Going to go ahead and merge. @vanroekel can ping me anytime to tweak the implementation.

@jgfouca jgfouca merged commit aa917ee into master Aug 16, 2017
@jgfouca jgfouca deleted the jgfouca/refactor_spare_nodes branch August 16, 2017 16:15
jgfouca added a commit that referenced this pull request Oct 17, 2017
Disable performance archiving if timing dir. cannot be written into

Currently, if the directory where performance data and associated
provenance data are to be saved already exists or cannot be created
or populated, then the job aborts. This change allows the job to proceed,
simply disabling the performance data archiving. This will allow the use of
default locations for saving the performance data that are legal for
users in the E3SM project but which will not be legal for E3SM users who
are not part of the project. These external users can define alternative
locations in their job cases.

BFB

Fixes #1816

* origin/worleyph/cime/e3sm_perf_archiving:
  Modifying warning message when archiving of timing data is disabled (4)
  Modifying warning message when archiving of timing data is disabled (3)
  Modifying warning message when archiving of timing data is disabled (2)
  Modifying warning message when archiving of timing data is disabled
  Make similar changes in _save_postrun_timing_acme function
  Disable performance archiving if timing dir. can not be written into
jgfouca added a commit that referenced this pull request Feb 23, 2018
Disable performance archiving if timing dir. cannot be written into

Currently, if the directory where performance data and associated
provenance data are to be saved already exists or cannot be created
or populated, then the job aborts. This change allows the job to proceed,
simply disabling the performance data archiving. This will allow the use of
default locations for saving the performance data that are legal for
users in the E3SM project but which will not be legal for E3SM users who
are not part of the project. These external users can define alternative
locations in their job cases.

BFB

Fixes #1816

* origin/worleyph/cime/e3sm_perf_archiving:
  Modifying warning message when archiving of timing data is disabled (4)
  Modifying warning message when archiving of timing data is disabled (3)
  Modifying warning message when archiving of timing data is disabled (2)
  Modifying warning message when archiving of timing data is disabled
  Make similar changes in _save_postrun_timing_acme function
  Disable performance archiving if timing dir. can not be written into
jgfouca added a commit that referenced this pull request Mar 13, 2018
Disable performance archiving if timing dir. cannot be written into

Currently, if the directory where performance data and associated
provenance data are to be saved already exists or cannot be created
or populated, then the job aborts. This change allows the job to proceed,
simply disabling the performance data archiving. This will allow the use of
default locations for saving the performance data that are legal for
users in the E3SM project but which will not be legal for E3SM users who
are not part of the project. These external users can define alternative
locations in their job cases.

BFB

Fixes #1816

* origin/worleyph/cime/e3sm_perf_archiving:
  Modifying warning message when archiving of timing data is disabled (4)
  Modifying warning message when archiving of timing data is disabled (3)
  Modifying warning message when archiving of timing data is disabled (2)
  Modifying warning message when archiving of timing data is disabled
  Make similar changes in _save_postrun_timing_acme function
  Disable performance archiving if timing dir. can not be written into
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants