Skip to content

Support multiple pools #27

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

Open
alexdremov wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Support multiple pools #27

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
94821d7
Change node-type argument to accept multiple values
alexdremov Mar 6, 2026
1595583
Clarify help text for --node-type argument
alexdremov Mar 6, 2026
a87ca44
Update csub.py
alexdremov Mar 6, 2026
19c81f7
Update csub.py
alexdremov Mar 6, 2026
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
6 changes: 3 additions & 3 deletions csub.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,7 +56,7 @@ def build_parser() -> argparse.ArgumentParser:
parser.add_argument("--secret-name", type=str, help="Override RUNAI_SECRET_NAME from the env file")
parser.add_argument("--pvc", type=str, help="Override SCRATCH_PVC from the env file")
parser.add_argument("--backofflimit", type=int, default=0, help="Retries before marking a training job as failed")
parser.add_argument("--node-type", type=str, choices=["", "v100", "h100", "h200", "default", "a100-40g"], default="", help="GPU node pool to target")
parser.add_argument("--node-type", nargs="*", type=str, choices=["v100", "h100", "h200", "default", "a100-40g"], default=[], help="GPU node pool(s) to target. Multiple values are accepted; the job will be scheduled on the first pool where it fits")
parser.add_argument("--host-ipc", action="store_true", help="Share the host IPC namespace")
parser.add_argument("--large-shm", action="store_true", help="Request a larger /dev/shm")
return parser
Expand Down Expand Up @@ -172,8 +172,8 @@ def build_runai_command(
cmd.append("--large-shm")

if args.node_type:
cmd.extend(["--node-pools", args.node_type])
if args.node_type in {"h200", "h100"} and not args.train:
cmd.extend(["--node-pools", ','.join(args.node_type)])
if any(i in {"h200", "h100"} for i in args.node_type) and not args.train:
cmd.append("--preemptible")

if distributed:
Expand Down
Loading