ML goodput gain Google reported from multi-tier checkpointing on one 35K-chip TPU v5p workload
+6.59% goodputobserved
| 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 reports a 6.59% increase in ML Goodput on one 35K-chip TPU v5p workload. The post does not say the gain is 6.59 percentage points absolute, and the result holds only for that workload, checkpoint cadence and storage tier. It is vendor-reported, not independently measured. Google's separate figure of almost $1M for a week-long 1K-VM GPU job is a hypothetical at an assumed 6.5% gain, not a measured saving. |
| As of | 2025 |
| Source | Google Cloud (multi-tier checkpointing) |
| Review | checking…review by 2027-06-17 · standard cadence |
| Recorded changes | last 2026-06-29 |
| Claim id | ml-goodput-gain-from-multi-tier-checkpointing |
Where the guide uses it
← Full numbers register — every date-stamped figure in the guide, with revision history.