Skip to content

Add MIG profile support for ml.p6-b300.48xlarge (Blackwell Ultra) #398

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
KeitaW wants to merge 1 commit into
base: main
Choose a base branch
from
+107 −0
Open

Add MIG profile support for ml.p6-b300.48xlarge (Blackwell Ultra) #398

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
106 changes: 106 additions & 0 deletions helm_chart/HyperPodHelmChart/charts/gpu-operator/config/default-mig-config.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -341,16 +341,122 @@ mig-configs:
"1g.24gb": 2
"2g.48gb": 1

# P6-B300 (Blackwell Ultra, 288GB HBM3e, ~269GB usable) profiles
# Profiles: 1g.34gb (x7), 1g.67gb (x4), 2g.67gb (x3), 3g.135gb (x2), 4g.135gb (x1), 7g.269gb (x1)
# Upstream ref: NVIDIA GPU Operator v25.3.0, device-filter 0x318210DE

all-1g.34gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.34gb": 7

all-1g.67gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.67gb": 4

all-2g.67gb:
- devices: all
mig-enabled: true
mig-devices:
"2g.67gb": 3

all-3g.135gb:
- devices: all
mig-enabled: true
mig-devices:
"3g.135gb": 2

all-4g.135gb:
- devices: all
mig-enabled: true
mig-devices:
"4g.135gb": 1

all-7g.269gb:
- devices: all
mig-enabled: true
mig-devices:
"7g.269gb": 1

mixed-1-3g.135gb-1-4g.135gb:
- devices: all
mig-enabled: true
mig-devices:
"3g.135gb": 1
"4g.135gb": 1

mixed-1-1g.34gb-1-2g.67gb-1-4g.135gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.34gb": 1
"2g.67gb": 1
"4g.135gb": 1

mixed-3-1g.34gb-1-4g.135gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.34gb": 3
"4g.135gb": 1

mixed-1-1g.34gb-1-2g.67gb-1-3g.135gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.34gb": 1
"2g.67gb": 1
"3g.135gb": 1

mixed-3-1g.34gb-1-3g.135gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.34gb": 3
"3g.135gb": 1

mixed-2-2g.67gb-1-3g.135gb:
- devices: all
mig-enabled: true
mig-devices:
"2g.67gb": 2
"3g.135gb": 1

mixed-2-1g.34gb-1-2g.67gb-1-3g.135gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.34gb": 2
"2g.67gb": 1
"3g.135gb": 1

mixed-4-1g.34gb-1-3g.135gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.34gb": 4
"3g.135gb": 1

mixed-1-1g.34gb-3-2g.67gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.34gb": 1
"2g.67gb": 3

mixed-3-1g.34gb-2-2g.67gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.34gb": 3
"2g.67gb": 2

mixed-5-1g.34gb-1-2g.67gb:
- devices: all
mig-enabled: true
mig-devices:
"1g.34gb": 5
"2g.67gb": 1
1 change: 1 addition & 0 deletions src/sagemaker/hyperpod/training/constants.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,6 +132,7 @@
'ml.p5e.48xlarge': ['mig-1g.18gb', 'mig-1g.35gb', 'mig-2g.35gb', 'mig-3g.71gb', 'mig-4g.71gb', 'mig-7g.141gb'],
'ml.p5en.48xlarge': ['mig-1g.18gb', 'mig-1g.35gb', 'mig-2g.35gb', 'mig-3g.71gb', 'mig-4g.71gb', 'mig-7g.141gb'],
'p6-b200.48xlarge': ['mig-1g.23gb', 'mig-1g.45gb', 'mig-2g.45gb', 'mig-3g.90gb', 'mig-4g.90gb', 'mig-7g.180gb'],
'ml.p6-b300.48xlarge': ['mig-1g.34gb', 'mig-1g.67gb', 'mig-2g.67gb', 'mig-3g.135gb', 'mig-4g.135gb', 'mig-7g.269gb'],
'ml.p6e-gb200.36xlarge': ['mig-1g.23gb', 'mig-1g.47gb', 'mig-2g.47gb', 'mig-3g.93gb', 'mig-4g.93gb', 'mig-7g.186gb'],
'ml.g7e.2xlarge': ['mig-1g.24gb', 'mig-2g.48gb', 'mig-4g.96gb'],
'ml.g7e.4xlarge': ['mig-1g.24gb', 'mig-2g.48gb', 'mig-4g.96gb'],
Expand Down
Loading