Even if a hard drive physically dies, data is often still salvageable. What are the chances of valuable data surviving a problem with an SSD though?
Every PC, business or personal, worthy of the name today has an SSD as its system drive. SSDs won the battle with HDDs years ago. At first, the SATA HDD offered faster speeds in a similar size to the HDD (about 500 MB/s in a 2.5-inch enclosure).
Later, the PCIe NVMe SSD emerged as the standard. Through that connection, transfer speeds of more than 1 GB/s are possible. Both connections are different, but the technology behind flash storage is the same.
SSDs are not only much faster than HDDs, they are theoretically more resilient since they contain no moving parts. That’s not to say they can’t crash. Given the role the SSD plays in a system, chances are that such a crash will impact crucial data.
That’s not an issue for our readers, of course, since they all have adequate immutable back-ups of everything, but what about the data of less proficient colleagues without a backup? With an SSD, is it also possible to recover data after a fatal crash, like it is with a hard drive?
The short answer is “yes.” The slightly longer answer is “Yes, but…”. SSDs are increasingly appearing on the recovery bench in specialized labs. Whether experts can recover data there depends mostly on the cause of the problem. We walk you through the scenarios.
A file has disappeared
First of all, it is interesting to know that the type of SSD does not really have an impact on recovery chances. SATA, SAS or NVMe: behind the connector always sits a controller with a collection of flash chips behind it, made up of SLC, TLC or MLC-NAND memory.
The simplest scenario is that of human error: you accidentally deleted your data, and that one important spreadsheet is unfortunately no longer in the recycle bin, either. In that case, the recovery process is quite similar to that of an HDD. A deleted file persists until it is overwritten on the storage medium. In other words, the bits and bytes of the spreadsheet can still be found on the SSD even after you hit the delete button. They are only untraceable by the OS.
Recovery of the data in that case is done with specialized software. This can be done in the lab, but you can basically do it yourself with the right tools. One thing is vital: stop using the SSD. The longer the SSD is active, the greater the chance that the deleted files will be overwritten and any chance of recovery disappears completely.
Trimming: killer of deleted files
With SSDs, there is an additional problem that HDDs do not have: trimming. The problem lies in the construction of the NAND flash memory that makes up the SSD. NAND cells that hold bits and bytes are grouped into a grid. Each row of such a grid is called a page, and pages are grouped into a block. However, SSDs cannot delete pages individually. Deletion is always done at the block level.
If you delete a file, the trim function of the SSD causes the block to be erased and its value set to 0. This makes it easier for the OS to write a new file to the empty block when needed, but has the side effect that the bits and bytes that make up your spreadsheet disappear even before they are overwritten. So if you press delete, your file still exists, but once trimming has taken place, the bits are gone.
Electrical damage
It becomes even more difficult when an SSD encounters electrical damage. For one thing, that could be a power surge that causes the circuits to burn out, but the reverse is also dangerous. If the power suddenly goes out, there is a risk of file corruption. Some (more professional) SSDs are protected against this.
It gets even more annoying when corruption occurs at the mapping table level. Files are written to an SSD across different NAND cells. The mapping table keeps track of which piece of file is where. If that table fails, the contents of the disk have to be puzzled back together, and that is not easy.
If the power suddenly goes out, there is a risk of file corruption.
If the damage after a power surge is limited to the controller, the problem is easily solved, especially on an older SSD. In the lab, technicians can then simply replace the faulty controller with a new one from the same manufacturer. That one can then basically get right to work on the underlying flash memory.
Reverse engineering
More complex problems, experts have to approach on a case-by-case basis. How an SSD stores data varies from manufacturer to manufacturer and even from model to model. Piecing data back together is only possible when you know how it was distributed across the SSD, and that often involves reverse engineering the controller firmware. After all, exactly how a controller works impacts the performance of an SSD. This is not something a manufacturer is willing to share with an external lab, but just such labs are your last hope when you want to save data.
The importance of the controller represent the difference with the greatest impact on data recovery, compared to HDDs. Whereas with a hard drive you still get soem data back as long as you can read the data on the platters, data is more likely to be unrecoverable on SSD’s.
Even when experts in the lab unsolder flash modules and manually read what’s on them them, they still don’t know how files were split up so they can’t just piece them together either.
If the drive is encrypted, the situation can really become hopeless. If the encryption key is lost with the SSD failure, the recovery team can see nothing more than random ones and zeros. Putting those back together again is an impossible task. Since the mid-2024s, Microsoft has been enabling BitLocker encryption by default on Windows 11, which is good news for security but bad for recovery capabilities.
Instability
Finally, the age of SSDs plays a role. The TLC, SLC or MLC-NAND cells that make up the drive’s memory have an expiration date. The more you use them, the faster they wear out. Over time, the cells become unstable and unreliable.
SSDs are designed with that in mind. On the one hand, they often have extra cells on board to replace old ones; on the other, they are designed for a predetermined amount of writes. Manufacturers build their SSDs to provide many years of trouble-free service, but at some point (just like an HDD) you run into wear and tear.
Your OS should tell you when an SSD is approaching retirement age. If you continue to use the drive anyway, you run the risk of corruption. Prevention is better than cure here. Even without a pop-up, there are signs you can tell it’s time to replace a drive. For example, copying files will suddenly take much longer. Whether data is salvageable from a SSD suffering from dementia, depends on a case-by-case basis. Data on the drives are much more vulnerable to partial damage than HDDs.
Over time, NAND cells become unstable and thus unreliable.
When just one memory chip on an SSD is corrupted, it is already too late and it becomes impossible to recover files. After all, they are written across all the flash chips. Once there is one unreadable chip, all files on the SSD are lost.
Backups
So recovering data from SSDs is not impossible, but significantly more challenging than with HDDs. In many more cases, the way an SSD works causes data to disappear forever. Even experts have little room to manoeuvre in in those case. If something goes wrong, turn off your system immediately and stop using the SSD, this way you increase the chances of successful recovery.
read also
In 3-2-1 to a successful backup strategy
Ideally, of course, you make sure your data does not depend on the survival of a single SSD. Things are physically more robust, but they can also break down. Something can always go wrong with a physical carrier. Therefore, always back up important files, preferably following the 3-2-1 rule.
This article originally appeared on July 19, 2019. It was updated on Feb. 26, 2025 with the latest information.