Meta RSC-1 versus RSC-2 observed job-failure counts normalized by allocated node-days
6.50 vs 2.34observed
| Value kind | observed — Reported measurements, counts, and specifications keep the precision and scope stated by their source; an exact specification is not treated as a range. |
|---|---|
| Scope | Two A100 research clusters with distinct workloads. Counts jobs using more than 128 GPUs; failures are NODE_FAIL jobs plus FAILED jobs attributed to a critical health check, divided by job runtime × allocated nodes. Not unique failed-FRU counts or a universally portable component failure intensity. |
| Caveat | Two named A100 research clusters and one event definition; not a universal component λ. |
| As of | 2024-10 |
| Source | Meta paper; use only against an identical event/exposure definition. · §III, nominal rf calculation before Analysis of ETTR Results |
| Review | checking…review by 2027-10-16 · standard cadence |
| Recorded changes | last 2026-10-16 · 3 revisions tracked |
| Claim id | failures-per-1-000-node-days-meta-rsc-1-vs-rsc |
Where the guide uses it
← Full numbers register — every date-stamped figure in the guide, with revision history.