SSD TBW and Wear Levelling, Explained: How Much Write Life Your Drive Really Has
Every SSD carries a quiet countdown. It does not tick in days but in writes: each flash cell can be erased and rewritten only so many times before it stops holding a charge reliably. Two ideas keep that countdown slow and readable — wear levelling, the drive’s own traffic management, and TBW, the published budget of total writes. Understanding both turns a scary specification into a number you can check against your own habits in about five minutes.
There is a second reason this matters more than it used to. A failing spinning drive often announced itself — clicking, grinding, growing louder. An SSD has no moving parts and makes no sound at any age, so the health percentage reported by a monitoring tool is not a nice extra; it is the only warning system the drive has.
Why writes wear a drive out
SSDs store data as electrical charges in NAND flash cells. Each program/erase cycle stresses the cell’s oxide layer, and over enough cycles the charge leaks too quickly to trust. Published explanations put the endurance of a cell at roughly 3,000 to 100,000 writes depending on the flash type, after which the cell ‘forgets’ new data. Reading data does not cause this wear in any meaningful amount — it is writes and erases that spend the budget.
Left unmanaged, everyday patterns would destroy a drive lopsidedly: the operating system rewrites the same logs, caches, and swap areas constantly, while a folder of photos may not change for years. The hot spots would die young while most of the drive sat nearly new.
Wear levelling: the drive spreads the damage
The controller’s answer is wear levelling: distributing write operations across all available cells so every block ages at roughly the same rate. When data must be rewritten, the controller preferentially writes to younger blocks and moves static data out of old, lightly-used blocks when needed, so no single area takes the whole workload. Supporting mechanisms help — a reserved over-provisioned area gives the controller spare blocks to work with, and write amplification (the extra internal writing caused by block erasing) is the inefficiency wear levelling has to fight. The result is that the drive ages as a whole, predictably, instead of failing at its busiest corner.
TBW and DWPD: the budget, stated two ways
TBW (terabytes written) is the total amount of data the manufacturer rates the drive to accept over its life before reaching its rated end of life. Scale examples make it concrete: a typical figure for a 250 GB SSD lies between 60 and 150 TBW, and a 1 TB drive may carry a rating around 560 TBW — in effect, completely filling and erasing the drive hundreds of times. DWPD (drive writes per day) restates the same budget as full-drive writes per day across the warranty period, calculated as TBW divided by warranty days and capacity; it is the enterprise shorthand, and most home users can ignore it once they have the TBW figure.
Two cautions keep the number honest. TBW is a rated, warrantied figure, not a self-destruct threshold: a drive can keep working past it, with risk rising rather than certainty arriving. And TBW says nothing about other failure modes — controller faults and electronics failures do not consult the write counter, which is why backups remain non-negotiable regardless of how healthy the percentage looks.
Work out your own write life, in four steps
- Find the rating. Look up your exact model’s TBW on the manufacturer’s specification page. Record capacity and TBW together; the same model line differs by size.
- Read what you have already written. Open a drive-health tool and find the total host writes (often labelled Total Host Writes, or shown as the data-written SMART figure) and the drive’s power-on time or the date you installed it.
- Divide writes by days in service. Example with round, checkable arithmetic: 24 TB written over 2 years is about 33 GB per day. Compare that with the published reality checks — reaching even 70 TBW takes roughly 190 GB written every day for a year, and at 20 GB a day a 240 TBW drive works out to over 30 years of write life.
- Project the remainder. (Rated TBW − writes so far) ÷ your daily writes = days of write life remaining at your current habits. Recheck the health percentage twice a year; if the daily rate jumps, find out why before worrying about the drive itself.
What success looks like: you end with three numbers on one line — rated TBW, total written so far, and your average GB per day — and a projected write life measured in years. For ordinary home and office use that projection is usually measured in decades, and the honest conclusion is that the drive will be replaced for capacity or speed long before writes kill it. Heavy video editing, constant virtual-machine churn, or a drive used as a download scratch disk are the realistic exceptions, and even then the fix is usually a bigger or higher-rated drive, not anxiety.
What actually shortens SSD life in practice
Keeping a drive nearly full works against it: with few free blocks, the controller must shuffle data more for every write, raising write amplification. Sustained professional workloads — multi-gigabyte renders written daily, database logging, continuous recording — are the usage patterns the TBW budget was invented for. Heat and sudden power loss cause their own damage independently of the write count. None of this changes the standing advice for this site: monitor the percentage, keep free space, and keep backups that do not depend on any single drive’s health, however reassuring its numbers look today.
Reading the health percentage without misreading it
Health tools compress all of this into a single percentage, and that number deserves a careful reading. On most SSDs the percentage is an estimate of remaining life based primarily on consumed write endurance, adjusted by the controller's own spare-block accounting. It is not a probability of failure this year, and it does not move linearly with your anxiety about it: a drive that drops from 100 to 97 percent in its first year and then holds for three years is behaving normally, because early writes - installing the operating system, applications, and your existing files - are the heaviest the drive will ever see in ordinary use.
Read the percentage alongside two companions. The first is total host writes, which converts the percentage back into the arithmetic of the previous section and tells you whether the wear matches your actual usage or something unexpected is writing constantly - a misbehaving log, a sync loop, or a drive being used as scratch space by software you forgot about. The second is the SMART warning set: reallocated or retired blocks, uncorrectable errors, and any overall status that has moved from healthy to caution. A falling percentage with clean warnings is a budgeting matter to plan around over years. A stable percentage with new uncorrectable errors is a backup-and-replace matter this week. The percentage tells you how much road is left; the warnings tell you whether the road itself is breaking, and the second answer always outranks the first.
FAQ
Does reading data wear out an SSD?
No meaningful wear comes from reads. The budget is spent by program/erase cycles — writes and erases degrade the cell’s oxide layer over time.
What happens when a drive reaches its TBW?
Nothing automatic. TBW is a rated and warrantied figure; drives commonly keep working past it, but the risk of failure rises, so a drive near its rating holding important data should be planned for replacement, and backed up regardless.
How do I check how much my SSD has written?
A drive-health tool reading SMART data shows total host writes for most drives, alongside a health percentage that already reflects wear levelling and consumed life.
Is a higher TBW worth paying for?
Only if your measured daily writes say so. Do the four-step calculation first: most home users project decades of write life on an ordinary rating, while heavy daily writers genuinely benefit from a higher one.