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

Fixing a problem with image restarts when using multiple processors #638

Merged
merged 3 commits into from Oct 3, 2016

Conversation

Projects
None yet
3 participants
@hhelgason
Contributor

hhelgason commented Oct 3, 2016

This PR fixes a problem with image restarts when using multiple processors. Edits were made to the vic_restore.c script to ensure that only the master node validates state file dimensions and coordinate variables. Multiprocessing was also added to a VIC Testing script.

Hörður Bragi Helgason and others added some commits Sep 29, 2016

Hörður Bragi Helgason
Committing a fix to vic_restore.c
Added an if statement so that when VIC is run on multiple nodes,
only the master node will validate state file dimensions and coordinate variables.
Merge pull request #1 from yixinmao/multiproc_init_state_fix
Use multiple processors for image restart tests

@bartnijssen bartnijssen merged commit 9a71edd into UW-Hydro:hotfix/5.0.1 Oct 3, 2016

1 check passed

continuous-integration/travis-ci/pr The Travis CI build passed
Details

hhelgason pushed a commit to hhelgason/VIC that referenced this pull request Oct 6, 2016

Hörður Bragi Helgason
Added release notes about PR UW-Hydro#638.
(Fixing a problem with image restarts when using multiple processors).

jhamman added a commit that referenced this pull request Oct 6, 2016

Added release notes about PR #638. (#640)
(Fixing a problem with image restarts when using multiple processors).

jhamman added a commit that referenced this pull request Jan 6, 2017

Feature/debug cesm driver (#642)
* update command line usage docs

* update release notes

* Fixed image driver history file name (#635)

* Fixed image driver history file name timetamp - now is timstep-beginning

* Added release notes about PR #635 (fixing image driver history filename
timestamp)

* Committing a fix to vic_restore.c

Added an if statement so that when VIC is run on multiple nodes,
only the master node will validate state file dimensions and coordinate variables.

* Use multiple processors for image restart tests

* dummy forcings are now created while initializing the model prior to communicating with the coupler

* removed decimal point for missing value for unsigned ints

* initial fix for getting state file date/time info

* updated fix for getting statefile time info from dmy_current (period-ending)

* Fixed a bug when calculating steps of forcing to skip  (#639)

* Fixed a bug when calculating steps of forcing to skip - should round to
integer first before converting to int type to prevent possible 1-timestep wrong offset

* Added bug fix release notes to PR#639

* fixed bug that passed fillvalue as double when it was supposed to be an int in gather_put_nc_field_int fctn

* Added release notes about PR #638. (#640)

(Fixing a problem with image restarts when using multiple processors).

* use appropriate fillvalue datatypes in gather_put_nc_field fctns

* add caseid to statefile name by passing case metadata in to vic_cesm_run function

* Revert "add caseid to statefile name by passing case metadata in to vic_cesm_run function because adding this logic to vic_cesm_init instead"

This reverts commit 7e28f13.

* assign caseid to statefile name in vic_cesm_init

* fix warnings for SNOW_BAND and AGGFREQ

* fixed typo in comment

* Added check to ensure mask variable in the input domain file is integer type (#645)

* Added check to ensure mask varible in domain file is integer type

* Cleaned up some printing lines

* Added get_nc_var_type.c

* Minor update of comments and ReleaseNotes

* Added domain file description in docs

* Small fix of table in docs

* Added ncdump -h results for domain file in docs

* Minor docs update

* fix typo in primary reference (#646)

* Fix bug make lastday (#647)

* Added check to ensure mask varible in domain file is integer type

* Cleaned up some printing lines

* Added get_nc_var_type.c

* Minor update of comments and ReleaseNotes

* Added domain file description in docs

* Small fix of table in docs

* Added ncdump -h results for domain file in docs

* Minor docs update

* The input arguments to function `make_lastday` are sometimes in a wrong
order - fixed the bug

* Added ReleaseNotes regarding PR#647

* Revert "Merge branch 'hotfix/5.0.1' of github.com:UW-Hydro/VIC into hotfix/5.0.1"

This reverts commit a9a202b, reversing
changes made to bbf7d9a.

* pressure was being converted to kPa, should be in Pa

* Revert "pressure was being converted to kPa, should be in Pa"

This reverts commit 72a56a5.

* pressure was being converted to kPa, should be in Pa

* convert pressure to kPa when reading it in rather than Pa since this was updated in VIC 5

* Revert "convert pressure to kPa when reading it in rather than Pa since this was updated in VIC 5"

This reverts commit e7192c5.

* fixed bug with using the correct local domain grid cells and units issues

* address PR review comments

* add change log entry
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment