[Fix] KVCache: enhance rdma auto-detection

[DwyaneShi]

Pull Request Description

Fallback to not using rdma auto-detection if pyverbs is not available on the runtime platform.

Related Issues

Resolves: #[Insert issue number(s)]

Important: Before submitting, please complete the description above and review the checklist below.

---

Contribution Guidelines (Expand for Details)

We appreciate your contribution to aibrix! To ensure a smooth review process and maintain high code quality, please adhere to the following guidelines:

Pull Request Title Format

Your PR title should start with one of these prefixes to indicate the nature of the change:

• [Bug]: Corrections to existing functionality
• [CI]: Changes to build process or CI pipeline
• [Docs]: Updates or additions to documentation
• [API]: Modifications to aibrix's API or interface
• [CLI]: Changes or additions to the Command Line Interface
• [Misc]: For changes not covered above (use sparingly)

Note: For changes spanning multiple categories, use multiple prefixes in order of importance.

Submission Checklist

• PR title includes appropriate prefix(es)
• Changes are clearly explained in the PR description
• New and existing tests pass successfully
• Code adheres to project style and best practices
• Documentation updated to reflect changes (if applicable)
• Thorough testing completed, no regressions introduced

By submitting this PR, you confirm that you've read these guidelines and your changes align with the project's contribution standards.

[gemini-code-assist]

Summary of Changes

Hello @DwyaneShi, 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 enhances the robustness of the KVCache's RDMA integration by implementing a graceful fallback mechanism. It allows the system to operate correctly even when the underlying pyverbs library (required for RDMA transport) is not installed, by either disabling RDMA auto-detection or requiring explicit device configuration. This improves the user experience by preventing crashes in environments without full RDMA support.

Highlights

• Conditional RDMA Transport Support: I've introduced a new global flag, HAS_RDMA_TRANSPORT_SUPPORT, in aibrix_kvcache/transport/__init__.py. This flag is set to True by default but will be set to False if importing the rdma module (which depends on pyverbs) fails due to an ImportError. This allows the system to dynamically determine if RDMA capabilities are available at runtime.
• Robust RDMA Auto-Detection: The logic for RDMA device auto-detection and explicit device configuration within infinistore.py is now conditionally executed based on the HAS_RDMA_TRANSPORT_SUPPORT flag. If RDMA transport is not supported on the platform, the system will bypass the RDMA-specific initialization. This prevents crashes in environments without pyverbs.
• Enforced Configuration for Non-RDMA Environments: When RDMA transport is not available, I've added an assertion that requires the AIBRIX_KV_CACHE_OL_INFINISTORE_VISIBLE_DEV_LIST environment variable to be explicitly set. This ensures that if auto-detection is not possible, users are forced to provide the necessary device configuration, preventing silent failures or unexpected behavior.

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

This pull request aims to handle cases where pyverbs is not installed by adding a check for RDMA transport support. The implementation in transport/__init__.py correctly detects the availability of the pyverbs library.

However, the usage of this check in l2/connectors/infinistore.py is flawed. It can allow the program to continue in a broken state if RDMA is configured without pyverbs, leading to less obvious errors later. I've left a critical comment with a suggestion to fail fast with a clear error message in this scenario.

Overall, the direction is good, but the fallback logic needs to be more robust.

[Jeffwan]

seems the primary change is

1. import package on demand
2. add feature flag HAS_RDMA_TRANSPORT_SUPPORT.

the core discovery logic has not changed a lot.

overall looks good to me