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.

Sign up for GitHub

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 force multi-node launcher mode #977

Merged
merged 2 commits into from
Apr 20, 2021
Merged

add option to force multi-node launcher mode #977

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2ba97d1
add option to force multi-node launcher mode
jeffra Apr 19, 2021
9857676
Merge branch 'master' into jeffra/launcher-update
jeffra Apr 20, 2021
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
7 changes: 6 additions & 1 deletion deepspeed/launcher/runner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -104,6 +104,11 @@ def parse_args(args=None):
help="(optional) pass launcher specific arguments as a "
"single quoted argument.")

parser.add_argument("--force_multi",
action="store_true",
help="Force multi-node launcher mode, helps in cases where user "
"wants to launch on single remote node.")

parser.add_argument("user_script",
type=str,
help="User script to launch, followed by any required "
Expand Down Expand Up @@ -304,7 +309,7 @@ def main(args=None):
# encode world info as base64 to make it easier to pass via command line
world_info_base64 = encode_world_info(active_resources)

multi_node_exec = len(active_resources) > 1
multi_node_exec = args.force_multi or len(active_resources) > 1

if not multi_node_exec:
deepspeed_launch = [
Expand Down