Onboard Tillicum: cluster doc, Slurm port, and first-login recon (#51/#70) - #91

Open

jonfroehlich wants to merge 7 commits into

mainfrom

docs/tillicum-onboarding

Commits on Jul 30, 2026

Add docs/tillicum.md: access, cost model, and the Slurm migration diff

jonfroehlich
and
claude
committed
9430266
Record the measured dataset size: 286 GB across ~911k files

jonfroehlich
and
claude
committed
315a672
Add the Tillicum training script and a first-login recon pass

jonfroehlich
and
claude
committed
5237e86
Save periodic checkpoints on Tillicum, and document what resume cannot inherit

jonfroehlich
and
claude
committed
fee620e

Commits on Jul 31, 2026

Fold in UW-IT's answers: SquashFS, /gpfs/scrubbed, and a safe enforced budget

jonfroehlich
and
claude
committed
3d12a7a
Regenerate the YOLO dataset on Tillicum, and prove it matches klone

jonfroehlich
and
claude
committed
f8c287e
Commit the Tillicum env-setup and smoke scripts, with what they measured

jonfroehlich
and
claude
committed
fa46946