feat(bigtable): introduce prime() method and load counting based on unary,streaming and ability to remove connection

[sushanb]

no-op from sushanb#1

[gemini-code-assist]

Summary of Changes

Hello @sushanb, 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 significantly refactors the Bigtable client's internal connection pooling mechanism. The primary goal is to enhance connection management by introducing a Prime method to warm up connections, refining load balancing through distinct tracking of unary and streaming requests, and implementing a robust graceful draining process for connections. These changes aim to improve the client's performance, reliability, and responsiveness by ensuring connections are efficiently utilized and gracefully managed throughout their lifecycle.

Highlights

• Connection Priming: Introduced a Prime() method for Bigtable connections to proactively warm them up using a PingAndWarm RPC, reducing initial request latency.
• Granular Load Tracking: Implemented separate atomic counters for in-flight unary and streaming requests (unaryLoad, streamingLoad) within each connection entry, allowing for more precise load balancing decisions.
• Graceful Connection Draining: Added functionality to gracefully drain connections when they are being removed or replaced. Connections are marked as draining and new requests are routed to other connections, while existing in-flight requests are allowed to complete within a configurable timeout.
• Enhanced Connection Pool Options: The NewBigtableChannelPool constructor now accepts a context.Context and new options to configure the pool with an app profile, instance name, feature flags metadata, logger, and meter provider.
• Improved Selection Strategies: Load balancing selection methods (Round Robin, Least In-Flight, Power of Two Least In-Flight) have been updated to skip connections that are in a draining state, ensuring requests are always sent to active connections.

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 by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

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 pull request 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 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. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

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 introduces significant enhancements to the Bigtable connection pool, including connection priming via a PingAndWarm RPC, fine-grained load tracking for unary and streaming calls, and a robust mechanism for connection draining and replacement. The changes are extensive, touching the connection pool implementation, its configuration, and its tests. The introduction of connEntry to manage individual connection state and the use of copy-on-write for updating the connection list are solid design choices for concurrent access. The test suite has been substantially improved with new tests for graceful draining, connection replacement, and comprehensive benchmarks.

My review focuses on a few areas for improvement. A key concern is a potential regression where the load balancing strategy is now hardcoded, removing previous configurability. I've also identified opportunities to simplify metadata handling and improve code clarity by replacing a magic number with a standard library constant. Finally, there's an unnecessary delay in one of the tests that could be removed to improve test execution speed.