DO NOT MERGE - debug

[robertgshaw2-redhat]

No description provided.

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

[mergify]

This pull request has merge conflicts that must be resolved before it can be
merged. Please rebase the PR, @robertgshaw2-redhat.

https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork

[gemini-code-assist]

Summary of Changes

Hello @robertgshaw2-redhat, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request is a debugging and development-focused effort, not intended for immediate merge. Its primary purpose is to introduce new tools and infrastructure to facilitate the testing and performance analysis of the prefill/decode (PD) architecture, specifically leveraging the NixlConnector. It also includes several enhancements to the NixlConnector itself to improve its efficiency and management of KV transfers.

Highlights

• New Sequential Benchmark: I've added a new Python script (benchmarks/benchmark_one_concurrent_req.py) designed for sequential benchmarking of LLM serving. This script processes requests one by one, measures Time to First Token (TTFT), Inter-token Latency (ITL), and End-to-End Latency (E2EL), and provides detailed statistical reporting. Note that the prefix cache reset logic within this benchmark is currently disabled.
• Development Workflow Enhancements: A new Justfile has been introduced in pd_justfile/Justfile to streamline the development and testing of the prefill/decode (PD) setup using the NixlConnector. This includes recipes for launching prefiller and decoder instances, a proxy server, sending test requests, and running the new sequential benchmark against both the proxy and directly against the prefiller.
• Dynamic Port Allocation Utility: A utility script (pd_justfile/port_allocator.py) has been added to dynamically allocate consistent port numbers based on the username and a base port. This helps prevent port collisions in shared development environments.
• NixlConnector Improvements: Significant changes have been made to vllm/distributed/kv_transfer/kv_connector/v1/nixl_connector.py. These include:

• Dynamic worker allocation for NixlWrapper based on whether the instance is a prefiller.
• Refined tracking of in-progress transfers, now including agent names and notification IDs.
• Modified transfer completion logic to send notifications upon full completion of a request's transfers.
• Implementation of chunked and batched KV transfers in _read_blocks (using a CHUNK_SIZE of 500) to potentially improve performance and parallelism by distributing transfers across workers.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in issue comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist is currently in preview and may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments to provide feedback.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

The code changes introduce several modifications to the NixlConnector in vllm, including adding a constant, adjusting worker initialization based on environment variables, modifying transfer handling, and implementing chunking for large transfers. The changes aim to improve performance and stability in distributed environments.