-
Notifications
You must be signed in to change notification settings - Fork 2.8k
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.
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
[Misc] Add OpenTelemetry support #4687
base: main
Are you sure you want to change the base?
Conversation
@rkooo567 can you please review this PR raised by Ronen. Thanks. |
yep will do tmrw! |
ace7fe2
to
899adef
Compare
Rebased to fix merge conflicts. |
ttft = metrics.first_token_time - metrics.arrival_time | ||
e2e_time = now - seq_group.metrics.arrival_time | ||
# attribute names are based on | ||
# https://github.com/open-telemetry/semantic-conventions/blob/main/docs/gen-ai/llm-spans.md |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the adoption of otel llm semantic convetion, this is great!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Great. also do we have a holistic view of different observability metrics in vLLM?
I see another effort here: #5041
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We have this issue: #3616 (comment)
Added a commit to pass all unit tests. |
vllm/engine/llm_engine.py
Outdated
e2e_time = now - seq_group.metrics.arrival_time | ||
# attribute names are based on | ||
# https://github.com/open-telemetry/semantic-conventions/blob/main/docs/gen-ai/llm-spans.md | ||
seq_span.set_attribute("gen_ai.response.model", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can I suggest using the constants from https://github.com/traceloop/openllmetry/blob/main/packages/opentelemetry-semantic-conventions-ai/opentelemetry/semconv/ai/__init__.py
?
We keep them up to date with the decisions we're making in the otel semantic conventions working group
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for your suggestion! I've added a dependency on opentelemetry-semantic-conventions-ai
and extended SpanAttributes
with the missing constants.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Code itself looks good to me. My main concern is that we should probably not include this by default? This requires to setup Jaeger (or other tracing tools), and this seems pretty heavy requirements for every user.
That said, it'd require 3 changes.
- remove the requirements from requirements-common.txt
- init_trace should try importing it and becomes no-op when packages are not downloaded.
- Fail the request if tracing is not enabled.
cc @simon-mo for thoughts!
return | ||
arrival_time_nano_seconds = int(seq_group.metrics.arrival_time * 1e9) | ||
|
||
with tracer.start_as_current_span( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
QQ: what's the e2e overhead of this API? Can you measure and lmk for prompt length 512 output length 128?
It needs to be opt-in (i think it is already?), similar to TGI:
|
This PR adds basic support for OpenTelemetry distributed tracing.
It includes changes to enable tracing functionality and improve monitoring capabilities.
I've also added a markdown with print-screens to guide users how to use this feature. You can find it here
FIX #3789
BEFORE SUBMITTING, PLEASE READ THE CHECKLIST BELOW AND FILL IN THE DESCRIPTION ABOVE
PR Checklist (Click to Expand)
Thank you for your contribution to vLLM! Before submitting the pull request, please ensure the PR meets the following criteria. This helps vLLM maintain the code quality and improve the efficiency of the review process.
PR Title and Classification
Only specific types of PRs will be reviewed. The PR title is prefixed appropriately to indicate the type of change. Please use one of the following:
[Bugfix]
for bug fixes.[CI/Build]
for build or continuous integration improvements.[Doc]
for documentation fixes and improvements.[Model]
for adding a new model or improving an existing model. Model name should appear in the title.[Frontend]
For changes on the vLLM frontend (e.g., OpenAI API server,LLM
class, etc.)[Kernel]
for changes affecting CUDA kernels or other compute kernels.[Core]
for changes in the core vLLM logic (e.g.,LLMEngine
,AsyncLLMEngine
,Scheduler
, etc.)[Hardware][Vendor]
for hardware-specific changes. Vendor name should appear in the prefix (e.g.,[Hardware][AMD]
).[Misc]
for PRs that do not fit the above categories. Please use this sparingly.Note: If the PR spans more than one category, please include all relevant prefixes.
Code Quality
The PR need to meet the following code quality standards:
format.sh
to format your code.docs/source/
if the PR modifies the user-facing behaviors of vLLM. It helps vLLM user understand and utilize the new features or changes.Notes for Large Changes
Please keep the changes as concise as possible. For major architectural changes (>500 LOC excluding kernel/data/config/test), we would expect a GitHub issue (RFC) discussing the technical design and justification. Otherwise, we will tag it with
rfc-required
and might not go through the PR.What to Expect for the Reviews
The goal of the vLLM team is to be a transparent reviewing machine. We would like to make the review process transparent and efficient and make sure no contributor feel confused or frustrated. However, the vLLM team is small, so we need to prioritize some PRs over others. Here is what you can expect from the review process:
action-required
label on the PR if there are changes required. The contributor should address the comments and ping the reviewer to re-review the PR.Thank You
Finally, thank you for taking the time to read these guidelines and for your interest in contributing to vLLM. Your contributions make vLLM a great tool for everyone!