Skip to content

Add n2-highmem and more n2-standard CPU device types - #1199

Merged
scaliby merged 1 commit into
AI-Hypercomputer:mainfrom
mridul-sahu:add-n2-highmem-cpu-device-types
Nov 4, 2026
Merged

scaliby merged 1 commit into
AI-Hypercomputer:mainfrom
mridul-sahu:add-n2-highmem-cpu-device-types

Conversation

@mridul-sahu

Copy link
Copy Markdown
Contributor

Title: Add n2-highmem and more n2-standard CPU device types

Adds CPU device types for n2-highmem-{2,4,8,16,32,48,64,96} and n2-standard-{8,16,48,96}, following #1128 (n2-standard-4). Entries are generated by get_cpu_system_characteristics_map (like the TPU helper) for VM counts 1, 2, 4, 5, 8, 10, 16, 20, 32, 40, 64, 80, 128, 160, 256, 320, 512, 640: powers of two and 5 × powers of two, at most 1,000 (GKE's nodes per node pool and zone). Existing CPU entries are unchanged.

Motivation: running many simulated TPU hosts per VM needs high-memory or mid-size machines and non-power-of-two VM counts.

Testing: unit tests for the helper and catalog lookups; make run-unittests 513 passed; pyink/pylint/mypy clean. Used end to end (as a runtime catalog patch) for cluster create and workload create with n2-highmem-8-320 and n2-highmem-4-640 on GKE.

@gemini-code-assist gemini-code-assist Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request expands the supported CPU machine types in XPK by adding several n2-highmem and n2-standard configurations. It introduces a helper function get_cpu_system_characteristics_map to dynamically generate CPU device types for various VM counts, updates the documentation and README to reflect these additions, and includes comprehensive unit tests to verify the new behavior. No review comments were provided, so I have no feedback to evaluate.

@scaliby
scaliby added this pull request to the merge queue Nov 4, 2026
Merged via the queue into AI-Hypercomputer:main with commit ddd1b0b Nov 4, 2026
29 of 30 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants