Fixes accelerator node counts and adds 3D topology support

[JyotinderSingh]

• The num_node formula used previously was inaccurate and caused failures when trying to provision multi-chip node pools.
• Also adds support for 3D topologies

[gemini-code-assist]

Summary of Changes

Hello @JyotinderSingh, 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 addresses critical issues in accelerator configuration by rectifying an inaccurate num_node formula, which previously hindered the provisioning of multi-chip node pools. Concurrently, it enhances the system's capabilities by introducing comprehensive support for 3D TPU topologies, particularly for v5p, ensuring more precise and flexible resource allocation for advanced machine learning workloads.

Highlights

• Accelerator Node Count Correction: The num_node formula for v2 and v3 TPUs was corrected, resolving inaccuracies that led to failures when provisioning multi-chip node pools.
• 3D Topology Support: Support for 3D topologies was added, specifically for v5p TPUs, by updating topology specifications and modifying the parsing regular expression.
• TPU Configuration Updates: The TPUS dictionary was revised to reflect accurate chips_per_VM values and updated topology mappings for various TPU types, including v2, v3, v5p, and v6e.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

• keras_remote/core/accelerators.py
  • Updated TpuSpec definitions for v2, v3, v5p, and v6e TPUs to reflect corrected chips_per_VM values and new topology mappings.
  • Added explanatory comments regarding topology references and the num_nodes calculation formula.
  • Modified the _TPU_TOPO_RE regular expression to support parsing of 3D topology strings (e.g., "2x2x2").
• keras_remote/core/accelerators_test.py
  • Updated existing test cases for v3 TPUs to align with the corrected chips_per_VM and topology values.
  • Added new test cases to validate the parsing and configuration of v5p TPUs with 3D topologies.

Activity

• No human activity has been recorded on this pull request yet.

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

The pull request correctly addresses the inaccurate node count formula and introduces support for 3D topologies (e.g., for TPU v5p). It also improves the accelerator string parser to handle these new topology formats. I've identified a likely typo in the machine type for TPU v3 and noted that some common configurations (8-chip) were removed for v2 and v3, which might be a regression.