Skip to content

feat: Increase number of concurrent NPs to be created to 100#1068

Merged
jamOne- merged 2 commits intoAI-Hypercomputer:mainfrom
jamOne-:xpk-fork-100-nps
Feb 20, 2026
Merged

feat: Increase number of concurrent NPs to be created to 100#1068
jamOne- merged 2 commits intoAI-Hypercomputer:mainfrom
jamOne-:xpk-fork-100-nps

Conversation

@jamOne-
Copy link
Copy Markdown
Collaborator

@jamOne- jamOne- commented Feb 20, 2026

Description

Creating 100 NPs at once is considered safe and can speed up cluster creation process significantly for huge clusters.

Issue

http://b/477123549

Testing

This change increases the default batch size for run_commands from 10 to 100.
This allows for more concurrent nodepool operations, which significantly
speeds up operations on large clusters (e.g., creating 288 cubes).

Fixes b/477123549
Reverts the default batch size in run_commands to 10 and explicitly sets
batch=100 for nodepool creation commands.

Fixes b/477123549
@jamOne- jamOne- requested a review from scaliby as a code owner February 20, 2026 13:37
@jamOne- jamOne- added the release-features features label Feb 20, 2026
@jamOne- jamOne- enabled auto-merge February 20, 2026 13:37
@jamOne- jamOne- changed the title Xpk fork 100 nps feat: Increase number of concurrent NPs to be created to 100 Feb 20, 2026
@jamOne- jamOne- added this pull request to the merge queue Feb 20, 2026
Merged via the queue into AI-Hypercomputer:main with commit 678fcfe Feb 20, 2026
23 checks passed
@jamOne- jamOne- deleted the xpk-fork-100-nps branch February 20, 2026 13:54
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants