Description

This PR introduces a throughput benchmarking framework for Promptflow Serve which introduces tests for the following scenarios.

• WSGI + Flask + DAG + Sync Nodes
• ASGI + FastApi + DAG + Async Nodes
• ASGI + FastApi + Flex Flow + Async Nodes

It uses a minimum example that has a flow as shown below. The scenario covers 3 parallelly executing nodes (i.e. simulating guardrails) and upon completion of those, one other node that calls an http endpoint (i.e. simulating LLM or custom API).

The tests are controlled by the configuration parameters in the benchmark/promptflow-serve/test_runner/settings.env file.

The makefile located at benchmark/promptflow-serve/makefile

All Promptflow Contribution checklist:

• The pull request does not introduce [breaking changes].
• CHANGELOG is updated for new features, bug fixes or other significant changes.
• I have read the contribution guidelines.
• Create an issue and link to the pull request to get dedicated review from promptflow team. Learn more: suggested workflow.

General Guidelines and Best Practices

• Title of the pull request is clear and informative.
• There are a small number of commits, each of which have an informative message. This means that previously merged commits do not appear in the history of the PR. For more information on cleaning up the commits in your PR, see this page.

Testing Guidelines

• Pull request includes test coverage for the included changes.

Contributions

This PR contains contributions from @david-ratcliffe and @swetha-sundar.