[DO NOT REVIEW] AI gateway rate limits design & quick POC

[harupy]

Related Issues/PRs

#xxx

What changes are proposed in this pull request?

mlflow gateway start --config-path examples/gateway/openai/config.yaml

curl -X 'POST' 'http://127.0.0.1:5000/gateway/chat/invocations' -H 'accept: application/json' -H 'Content-Type: application/json' -d '{
  "max_tokens": 10,
  "messages": [
    {
      "role": "user",
      "content": "hello"
    }
  ]
}'

How is this PR tested?

• Existing unit/integration tests
• New unit/integration tests
• Manual tests

Does this PR require documentation update?

• No. You can skip the rest of this section.
• Yes. I've updated:
  • Examples
  • API references
  • Instructions

Release Notes

Is this a user-facing change?

• No. You can skip the rest of this section.
• Yes. Give a description of this change to be included in the release notes for MLflow users.

What component(s), interfaces, languages, and integrations does this PR affect?

Components

• area/artifacts: Artifact stores and artifact logging
• area/build: Build and test infrastructure for MLflow
• area/docs: MLflow documentation pages
• area/examples: Example code
• area/gateway: AI Gateway service, Gateway client APIs, third-party Gateway integrations
• area/model-registry: Model Registry service, APIs, and the fluent client calls for Model Registry
• area/models: MLmodel format, model serialization/deserialization, flavors
• area/recipes: Recipes, Recipe APIs, Recipe configs, Recipe Templates
• area/projects: MLproject format, project running backends
• area/scoring: MLflow Model server, model deployment tools, Spark UDFs
• area/server-infra: MLflow Tracking server backend
• area/tracking: Tracking Service, tracking client APIs, autologging

Interface

• area/uiux: Front-end, user experience, plotting, JavaScript, JavaScript dev server
• area/docker: Docker use across MLflow's components, such as MLflow Projects and MLflow Models
• area/sqlalchemy: Use of SQLAlchemy in the Tracking Service or Model Registry
• area/windows: Windows support

Language

• language/r: R APIs and clients
• language/java: Java APIs and clients
• language/new: Proposals for new client languages

Integrations

• integrations/azure: Azure and Azure ML integrations
• integrations/sagemaker: SageMaker integrations
• integrations/databricks: Databricks integrations

How should the PR be classified in the release notes? Choose one:

• rn/none - No description will be included. The PR will be mentioned only by the PR number in the "Small Bugfixes and Documentation Updates" section
• rn/breaking-change - The PR will be mentioned in the "Breaking Changes" section
• rn/feature - A new user-facing feature worth mentioning in the release notes
• rn/bug-fix - A user-facing bug fix worth mentioning in the release notes
• rn/documentation - A user-facing documentation change worth mentioning in the release notes

[github-actions]

Documentation preview for 15ce328 will be available here when this CircleCI job completes successfully.

More info

• Ignore this comment if this PR does not change the documentation.
• It takes a few minutes for the preview to be available.
• The preview is updated when a new commit is pushed to this PR.
• This comment was created by https://github.com/mlflow/mlflow/actions/runs/6529357779.

[harupy]

Global limit vs. per-route limit.

[serena-ruan]

per-route limit

[harupy]

• https://github.com/laurentS/slowapi
• https://github.com/laurentS/slowapi/blob/master/docs/examples.md#examples

[harupy]

Do we need multiple limits?

[harupy]

@TomeHirata Would you be interested in working on this?

[TomeHirata]

@harupy Yes, is there any ticket for this task? If so, please assign this task to me!

[TomeHirata]

@harupy I read the doc and understood the overall direction. Before starting the implementation, can I ask if there is a way for me to see this link?
I also saw mlflow.gateway.get_limits is deprecated. Should I reuse the method? or recreate a new one?

[harupy]

can I ask if there is a way for me to see this link?

Sorry, this is a link to our internal repo.

I also saw mlflow.gateway.get_limits is deprecated. Should I reuse the method? or recreate a new one?

mlflow.gateway has been deprecated and replaced by mlflow.deployments. We no longer have get_limits. The rate limit info should be included in the response of get_endpoint. I can take care of that part.

[TomeHirata]

Thank you for the reply. So what did the link refer to? Is it the same as the example code in the design doc?
Regarding the scope of the change, can I understand the necessary changes are as follows? Anything I'm missing?

1. Read the rate limit configuration from the endpoint yaml
2. Implement the rate limiting with slowapi
3. Include the rate limit info in the response of get_endpoint

[harupy]

Is it the same as the example code in the design doc?

Yes :)

Regarding the scope of the change, can I understand the necessary changes are as follows? Anything I'm missing?

Yes, that's correct!

[TomeHirata]

Thank you, let me start working on this shortly. Should I take over this PR? or create a new one?
->Update: I cannot take over this PR since this is not my repo, let me create a new one.