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

Add option to run multiple jobs in a row #71

Open
wants to merge 3 commits into
base: gcp
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
73 changes: 45 additions & 28 deletions buildstockbatch/gcp/gcp.py
Original file line number Diff line number Diff line change
Expand Up @@ -1102,12 +1102,13 @@ def main():
GcpBatch.run_combine_results_on_cloud(gcs_bucket, gcs_prefix, results_dir, do_timeseries)

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe we move this into develop at this point instead of gcp?

else:
parser = argparse.ArgumentParser()
parser.add_argument("project_filename")
parser.add_argument("project_filenames", help="Comma-separated list of project YAML files to run.")

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

maybe removing the help part from here as it is in the argument section as well?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not sure what you mean... But I like the help string here because it shows up when you run buildstock_gcp --help:

...
positional arguments:
  project_filenames  Comma-separated list of project YAML files to run.
  job_identifiers    Comma-separated list of job IDs to use. Optional override of gcp.job_identifier in your project file. Max 48 characters.
...

parser.add_argument(
"job_identifier",
"job_identifiers",
nargs="?",
default=None,
help="Optional override of gcp.job_identifier in your project file. Max 48 characters.",
help="Comma-separated list of job IDs to use. "
"Optional override of gcp.job_identifier in your project file. Max 48 characters.",
)
group = parser.add_mutually_exclusive_group()
group.add_argument(
Expand Down Expand Up @@ -1148,34 +1149,50 @@ def main():
else:
logger.setLevel(logging.INFO)

# validate the project, and if --validateonly flag is set, return True if validation passes
GcpBatch.validate_project(os.path.abspath(args.project_filename))
if args.validateonly:
return True
project_filenames = args.project_filenames.split(",")
n_projects = len(project_filenames)
job_IDs = len(project_filenames) * [None]
if args.job_identifiers:
job_IDs = args.job_identifiers.split(",")
if len(job_IDs) != n_projects:

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this capturing the project_id issue? like trying to see what is the likelihood of this error happening?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is just checking that if you give a list of project files and a list of IDs, they're the same length.

raise ValidationError(
f"job_identifiers contains {len(args.job_identifiers.split(','))} IDs, "
f"but project_filenames contains {n_projects} files"
)

batch = GcpBatch(args.project_filename, args.job_identifier, missing_only=args.missingonly)
if args.clean:
batch.clean()
return
if args.show_jobs:
batch.show_jobs()
for project_filename, job_ID in zip(project_filenames, job_IDs):
logger.info(f"----------Validating {project_filename}{f' ({job_ID})' if job_ID else ''}----------")
# validate the project, and if --validateonly flag is set, return True if validation passes
GcpBatch.validate_project(os.path.abspath(project_filename))

if args.validateonly:
return
elif args.postprocessonly:
if batch.check_for_existing_jobs(pp_only=True):
return
batch.build_image("gcp")
batch.push_image()
batch.process_results()
else:
if batch.check_for_existing_jobs():
return
if not args.missingonly:
batch.check_output_dir()

batch.build_image("gcp")
batch.push_image()
batch.run_batch()
batch.process_results()
for project_filename, job_ID in zip(project_filenames, job_IDs):

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Project-Job Pair Processing continues all the way to post-processing and then the next project-job pair is picked up or once the results are avilable in gcs, this will pick up the next job?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Right now, this is just running one job completely (waiting for post-processing to finish), then starting the next one. We could potentially start the second job while the first is in post-processing, but I'm not doing that here.

logger.info(f"----------Starting {project_filename}{f' ({job_ID})' if job_ID else ''}----------")
batch = GcpBatch(project_filename, job_ID, missing_only=args.missingonly)
if args.clean:
batch.clean()
continue
if args.show_jobs:
batch.show_jobs()
continue
elif args.postprocessonly:
if batch.check_for_existing_jobs(pp_only=True):
continue
batch.build_image("gcp")
batch.push_image()
batch.process_results()
else:
if batch.check_for_existing_jobs():
continue
if not args.missingonly:
batch.check_output_dir()

batch.build_image("gcp")
batch.push_image()
batch.run_batch()
batch.process_results()


if __name__ == "__main__":
Expand Down
4 changes: 4 additions & 0 deletions docs/run_sims.rst
Original file line number Diff line number Diff line change
Expand Up @@ -167,6 +167,10 @@ You can optionally override the ``job_identifier`` from the command line
(unless you delete a previous job with the ``--clean`` option), so this option makes it easier to
quickly assign a new ID with each run without updating the config file.

You can also run multiple projects in a row by specifying multiple project files (and optionally multiple IDs):
``buildstock_gcp project_file1,yaml,project_file2.yaml job_id1,job_id2``.
This will validate all the projects upfront, then run them sequentially.


Retry failed tasks
..................
Expand Down
Loading