1 Gi OOM'd during a real chain step. Each forecast hour loads both a surface and pressure-level HRRR GRIB2 (~40 MB combined), decodes via wgrib2 subprocess (which allocates its own buffers), scores 92k grid points × 23 bands (~2.1M intermediates), and writes 23 score files. 2 Gi absorbs the spike; still ~3× smaller than the Elixir-per-pod budget we're replacing. |
||
|---|---|---|
| .. | ||
| deployment-backfill.yaml | ||
| deployment-grid-rs.yaml | ||
| deployment.yaml | ||
| flux.yaml | ||
| kustomization.yaml | ||
| metrics-service.yaml | ||
| namespace.yaml | ||
| rbac.yaml | ||
| service.yaml | ||