checkpoint restore latency with multi-tier checkpointing on Google's 35K-chip TPU v5p workload — one clock inside MTTR
restore <1 minobserved
| 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 | Google’s named 35K-chip TPU v5p checkpoint restore measurement is less than one minute. It excludes detection, fencing, allocation, worker launch and validation; not an end-to-end MTTR. |
| As of | 2025-06-16 |
| Source | Google Cloud, Save early and often with multi-tier checkpointing to optimize large AI training jobs, 16 June 2025. · Multi-tier checkpointing architecture; restore operation on the named TPU v5p workload, separate from the complete recovery path. |
| Review | checking…review by 2026-09-07 · standard cadence |
| Recorded changes | last 2026-09-16 · 3 revisions tracked |
| Claim id | mttr-achievable-with-multi-tier-checkpointing |
Where the guide uses it
← Full numbers register — every date-stamped figure in the guide, with revision history.