jgrusewski 1931d93d66 feat(ci): add 2×H100 GPU pool for multi-GPU training ...

Pool ci-training-h100x2 (H100-2-80G, 160GB VRAM total).
MultiGpuConfig::detect() in ml crate auto-enables data parallelism.
Tags: kapsule,h100x2. Autoscale 0→1.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-03-04 11:32:26 +01:00

..

docker

ci: eliminate foxhunt-node-builder image, use node:22-slim directly

2026-02-28 18:40:51 +01:00

k8s

feat(ci): add 2×H100 GPU pool for multi-GPU training

2026-03-04 11:32:26 +01:00

live/production

chore(infra): remove dead Cockpit TF, CI-CD dashboards, stale Grafana configs

2026-03-02 15:13:40 +01:00

modules

infra: expose monitoring service via monitor.fxhnt.ai Tailscale proxy

2026-03-02 22:56:40 +01:00

scripts

feat(infra): enable MinIO TLS with self-signed CA for in-cluster HTTPS

2026-03-01 11:56:06 +01:00