Skip to content

Fix metrics and allow disable block manager v2 #1299

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
Nero10578 wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Fix metrics and allow disable block manager v2 #1299

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
8152f11
scheduler changes to allow LoRA + Chunked Prefill
Nero10578 Apr 14, 2025
c9bd904
allow block-manager-v2-disable and fix metrics negative value error
Nero10578 Apr 15, 2025
73b7409
Merge branch 'aphrodite-engine:main' into Fix-metrics-and-disable-blo…
Nero10578 Apr 15, 2025
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
5 changes: 2 additions & 3 deletions aphrodite/engine/args_tools.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -619,9 +619,8 @@ def add_cli_args(parser: FlexibleArgumentParser) -> FlexibleArgumentParser:
parser.add_argument(
'--use-v2-block-manager',
default=EngineArgs.use_v2_block_manager,
action='store_true',
help='Use BlockSpaceMangerV2. By default this is set to True. '
'Set to False to use BlockSpaceManagerV1')
type=lambda x: (str(x).lower() in ['true', '1', 'yes']),
help='Use BlockSpaceManagerV2 (default: True). Set explicitly to False to use BlockSpaceManagerV1')
parser.add_argument(
"--scheduler-delay-factor",
"-sdf",
Expand Down
3 changes: 3 additions & 0 deletions aphrodite/engine/metrics.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -468,6 +468,9 @@ def _log_gauge(self, gauge, data: Union[int, float]) -> None:

def _log_counter(self, counter, data: Union[int, float]) -> None:
# Convenience function for logging to counter.
if data < 0:
logger.warning(f"Attempted to increment counter with negative value: {data}")
return
counter.labels(**self.labels).inc(data)

def _log_counter_labels(self, counter, data: CollectionsCounter,
Expand Down