Measured counter trend
SSD Remaining Endurance Planner
Use two write-counter readings to plan a review threshold and replacement preparation date without claiming a failure date.
Measured endurance trend
Decision guide
Use measured writes to schedule attention, not to forecast failure
What this tool helps you decide
This planner turns two observed counter readings into a daily write-rate trend and a date at which a chosen TBW threshold may be reached. It helps set monitoring and replacement-preparation milestones. It deliberately does not name a physical failure date. You can use a plain TB-written counter or NVMe Data Units Written, where one unit equals 512,000 bytes. The selected threshold is a management policy, not a statement about drive health.
Before you start
Record the exact counter name, unit, source, date, and rated TBW from the SSD documentation. Take readings far enough apart to smooth a quiet day or a backup spike. Check that both readings came from the same drive and same counter definition. Enter SMART Percentage Used only as an optional comparison, because interpretation varies by vendor and interface. Choose a threshold and lead time that allow procurement, migration, validation, and recovery testing.
How the calculation works
NVMe data units convert to decimal TB by multiplying units by 512,000 bytes and dividing by one trillion. The measured rate is current TB minus previous TB, divided by days. Future daily writes grow that rate by the stated percentage. Threshold TB equals rated TBW times the threshold percentage. Remaining TB is threshold TB minus current writes. Days to threshold divide remaining TB by future daily TB; the preparation date subtracts the lead time.
How to interpret the results
A zero measured rate means there is no evidence for a threshold-date projection, not that the drive has infinite endurance. A counter lower than the previous reading is flagged as a reset or unit mismatch. If current writes exceed rated TBW, the planner continues to report the condition as a warning. Compare the rate with upcoming backup, cache, VM, and logging changes before acting. A date is best used to schedule another measurement and a replacement decision review.
Worked example
Suppose a 600 TBW, 1,000 GB drive rises from 100 TB to 115 TB in 30 days. Its measured rate is 0.5 TB per day, or 500 GB per day and 0.5 DWPD. With 10 percent expected growth and an 80 percent threshold, remaining threshold allowance is 365 TB. The model estimates about 664 days to that threshold. A 60-day preparation lead moves the procurement review earlier.
Assumptions and limitations
Counter availability, units, reset behavior, and Percentage Used meaning differ by vendor and interface. The model assumes a straight write rate with one adjustable growth percentage; workloads can change abruptly after migrations, retention changes, cache activation, or new applications. TBW and SMART values do not establish a physical failure date. Do not replace backup verification, alerting, error review, or vendor support procedures with this calculation. Always preserve a tested recovery path before maintenance.
Next steps
Re-run this tool after the next observation interval and compare the rate rather than relying on one sample. Use the TBW & DWPD Converter to normalize a new candidate rating, then use the Drive Replacement Reserve Calculator to fund the operational response. Record observed exceptions such as a counter reset in the maintenance log.