Skip to content

fix: fixed Planner division by zero errors in sglang backend #3113

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
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

fix: fixed Planner division by zero errors in sglang backend #3113

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5a1f4f9
fix: fixed Planner division by zero errors in sglang backend
GavinZhu-GMI Sep 18, 2025
f7154d5
fix: fixed format for precommit
GavinZhu-GMI Sep 18, 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
34 changes: 28 additions & 6 deletions components/planner/src/dynamo/planner/utils/planner_core.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -372,16 +372,38 @@ async def make_adjustments(self):
expect_ttft = self.prefill_interpolator.interpolate_ttft(
self.last_metrics.isl
)
self.p_correction_factor = self.last_metrics.ttft / expect_ttft
if expect_ttft > 0:
self.p_correction_factor = self.last_metrics.ttft / expect_ttft
else:
logger.warning(
f"Expected TTFT is {expect_ttft}, using default correction factor 1.0"
)
self.p_correction_factor = 1.0
# for ITL, we expect the correction factor to be close to 1
if len(self.d_endpoints) > 0:
concurrency = (
self.last_metrics.num_req # type: ignore
/ len(self.d_endpoints)
* self.last_metrics.request_duration # type: ignore
/ self.args.adjustment_interval
)
else:
logger.warning(
"No decode workers available, using default concurrency of 1.0"
)
concurrency = 1.0

expect_itl = self.decode_interpolator.interpolate_itl(
concurrency=self.last_metrics.num_req # type: ignore
/ len(self.d_endpoints)
* self.last_metrics.request_duration # type: ignore
/ self.args.adjustment_interval,
concurrency=concurrency,
context_length=self.last_metrics.isl + self.last_metrics.osl / 2, # type: ignore
)
self.d_correction_factor = self.last_metrics.itl / expect_itl
if expect_itl > 0:
self.d_correction_factor = self.last_metrics.itl / expect_itl
else:
logger.warning(
f"Expected ITL is {expect_itl}, using default correction factor 1.0"
)
self.d_correction_factor = 1.0
logger.info(
f"Correction factors: TTFT: {self.p_correction_factor:.3f}, ITL: {self.d_correction_factor:.3f}"
)
Expand Down
Loading