Solid state drives (SSDs) are becoming increasingly common in computers and other devices. Unlike traditional hard disk drives (HDDs), SSDs have no moving parts and instead store data in flash memory chips. This makes them faster, lighter, and less prone to mechanical failure than HDDs. However, SSDs are not infallible, and they can and do fail. When an SSD goes bad, there are some key things that tend to happen.
SSD Failure Modes
There are two main ways that SSDs can fail:
- Logical failure – Where the drive’s firmware gets corrupted or there are issues with the flash memory chips retaining data properly. This may result in data corruption or the drive becoming undetectable to the operating system.
- Complete failure – Where the electronics on the SSD’s circuit board stop working completely. This renders the drive totally unusable.
In many cases, an SSD will exhibit warning signs before completely failing. For example, logical failures may only affect certain sectors initially. The drive may start slowing down or showing much slower read/write speeds. Full failure of an SSD can happen suddenly without warning though.
Symptoms of SSD Failure
Here are some common symptoms that indicate an SSD may be failing or degrading:
- Errors and hangs during operation – Such as transfer freezes, input/output errors, or drives becoming totally unresponsive.
- Failure to boot – SSD is detected in BIOS but drive does not load OS. Or sometimes drive is not even detected during boot.
- Bad sectors – Parts of the drive become inaccessible. Any data stored there cannot be read.
- Reduced performance – Most notably slow read/write speeds compared to when the drive was new.
- Unusual noises – Clicking or buzzing sounds from the SSD, which should normally operate silently.
- SMART errors – Tools that can read a drive’s SMART data will detect reallocated sectors or read errors.
- Filesystem corruption – Such as invalid data being present or files/folders going missing. Checksums will not match.
One or more of the above symptoms indicates the drive electronics or flash memory are degrading. While warning signs may precede total failure, SSDs can also suddenly stop working without any signs of issues.
Recovering Data from a Failed SSD
If an SSD has failed completely and is no longer detected or accessible, data recovery becomes very difficult for the average user. Specialized data recovery companies use techniques like:
- Opening the SSD in a clean room environment to access the flash memory chips directly and copy their contents.
- Repairing or replacing damaged electronic components on the circuit board.
- Using advanced software to reconstruct data and filesystems.
Such data recovery can cost hundreds or thousands of dollars in many cases. And there is no guarantee all data can be salvaged from a failed drive. The techniques rely on the physical flash chips retaining data properly despite electronic failures elsewhere.
In limited cases where the drive is partially functional, some data may be recoverable by end users. For example, using recovery software to make a clone of the drive before attempting file extraction. But technical expertise is advised when attempting to recover anything from a failing SSD.
Causes of SSD Failure
There are a number of potential causes for SSDs to fail or start degrading, including:
- Write cycle exhaustion – SSDs can only sustain a finite number of writes before cells wear out. When fresh cells run out, drive failure occurs.
- Electrical faults – Power surges, controllers malfunctioning or short circuits on the board can all impact operation.
- Overheating – Heat buildup causes more rapid degradation of the flash memory and electronics.
- Firmware bugs – Bugs in the SSD’s controller firmware can cause glitches or crashes.
- Physical damage – Impacts can destroy components on the SSD or interrupt circuit board traces.
- Manufacturing defects – Imperfections introduced during the manufacturing process later lead to problems.
Most modern SSDs have lifespans in the years or decades for typical consumer usage. But heavy workloads like in a server environment, excessive writes/rewrites or inadequate cooling accelerate wear on the drive. This shortens the usable life considerably in some cases.
Preventing SSD Failure
While SSD failure is inevitable at some point, there are ways to help extend the usable life of a drive:
- Use the SSD properly within temperature limits and avoid overheating issues.
- Do not constantly write/rewrite data to all areas of the drive – this causes write amplification which wears out cells.
- Keep the drive firmware updated to latest stable release from the manufacturer.
- Monitor and keep spare space available. Filling an SSD completely causes more write amplification.
- Use drive cloning or imaging to take backups in case of failure.
- Handle SSD carefully to avoid physical damage from drops or shocks.
Enterprise SSDs designed for heavy workloads also employ techniques like better wear leveling algorithms, capacitors to backup writes in-progress, and single-level cell flash for longer lifespan. These come at a higher cost however.
Replacing a Failed SSD Under Warranty
Most SSDs come with a 3 to 5 year limited warranty from the manufacturer against defects and premature failure. If the drive stops working within that period, replacement options depend on the warranty terms but may include:
- Returning to the original place of purchase for a refund or exchange.
- Getting a replacement drive shipped from the manufacturer if registered/validated.
- Sending the failed drive back to the manufacturer for replacement.
Proof of purchase, serial numbers, error logs and other documentation are usually required under warranty processes. The replacement SSD may be new or refurbished depending on policy. Though manufacturers reserve the right to refuse replacement if the failure was due to user negligence or improper usage.
It is important to validate the warranty status and register SSDs as required soon after purchase. Return windows typically range from 15 to 90 days after the purchase date. Availing replacement under warranty also often voids further warranty on the replacement drive.
Buying a Replacement SSD
Once the warranty period has expired or replacement under warranty is not possible, the only option is to purchase a replacement SSD. Some points to keep in mind:
- Check compatibility – Choose an SSD with the correct physical form factor and interface for your device.
- Matching capacity – Get a drive with equal or higher storage capacity as the old SSD.
- Performance needs – Consider read/write speeds and use case when selecting SSD model.
- Budget – SSD prices vary widely based on brand, capacity, performance and features.
- Reliability – Check warranty, ratings and failure rate data before purchasing.
Migrating data from the old drive will require making clones, imaging or using disk mirroring software. Be prepared to reinstall operating systems and software on the new SSD as well.
Disposing of a Failed SSD Safely
Once an SSD has completely failed and been replaced, it should be disposed of carefully because:
- It contains sensitive data – This data could be retrieved by someone else if you throw the drive away carelessly.
- It contains toxic materials – There are metals and chemicals inside that should not end up in regular waste.
Some ways to safely dispose of a dead SSD:
- Physically destroy the drive – Drilling holes through the flash chips makes data unrecoverable.
- Perform a secure erase – Special software can overwrite the entire drive with meaningless data.
- Use a data destruction service – Many companies securely destroy and recycle old drives.
- Check for recycler accepting electronics – Some recyclers can extract and dispose of components safely.
Once satisfied the data cannot be recovered, the rest of the SSD components can be recycled. The enclosure and circuit board contain aluminum, copper, gold and various metals. The flash memory, capacitors and processor contain toxic substances however.
Recovering Data from a Failed SSD
When an SSD fails unexpectedly, the most immediate concern is recovering any important data that was stored on it. Unfortunately, traditional recovery methods rarely work on dead SSDs.
Unlike hard disk drives, recovering data directly from the NAND flash chips inside an SSD requires specialized skills and equipment. Some professional data recovery techniques include:
- Opening the SSD casing in a dust-free cleanroom to access the internal circuit board.
- Disassembling the boards and physically removing the flash memory chips.
- Using specialized adapters to connect the flash chips to a computer.
- Imaging the raw contents of the connected chips.
- Analyzing the images to identify recoverable partitions and files.
This requires expertise in electronics, specialized software tools and dust-free cleanroom facilities. Without proper precautions, attempts to open up a failed SSD could further damage the sensitive components inside.
Unfortunately, DIY recoveries are generally not possible on dead SSDs. But there are professional SSD data recovery services that may be able to recover data in some circumstances. They use similar techniques to those listed above, conducted in sophisticated cleanrooms.
The chances of recovering an SSD’s data depends on factors like:
- The cause and type of failure (e.g. electronic, firmware, physical damage).
- The make, model and age of the SSD.
- How much of the flash memory chips are still intact and functional.
- Whether both the SSD controller and flash chips have failed.
In a best case, full recovery may be possible by imaging the SSD’s flash chips. In worse cases, only fragments of files can be recovered from partially failed chips. And in some situations, even specialized cleanroom recovery cannot salvage any data.
SSD data recovery services typically charge a diagnostic fee upfront, followed by a larger fee if any data can be successfully recovered. Rates depend on the drive capacity, complexity and desired turnaround time. The costs could easily run over $1000 in many cases.
To avoid getting in this situation, it’s critical to maintain good backups of important data stored on any SSDs. Backups to separate external drives protect against data loss when internal SSDs fail unexpectedly.
Troubleshooting SSD Problems
Before concluding an SSD has completely failed, it is worth running through some troubleshooting steps in case it is a recoverable problem:
- Try a different cable/port – Use a known good SATA cable and connect to another SATA port if possible.
- Check for loose connections – Reseat SATA cables and power cables leading to the SSD.
- Update SSD firmware – A firmware bug may sometimes be causing detected issues.
- Try initializing/reformatting – This can fix filesystem errors and force remapping of bad sectors.
- Test with diagnostics software – Such as the drive maker’s own utilities to analyze health.
- Try on a different machine – The issues may be due to specific computer compatibility.
If the SSD is still not visible, exhibits problems or continues deteriorating, then it likely requires advanced repair or professional data recovery. Contact the manufacturer for warranty replacement if still eligible.
For systems that depend on an SSD for booting, it may be possible to clone the faulty drive before complete failure using disk imaging software. This preserves the data for recovery.
SSD Failure Rate Statistics
Studies looking at large samples of SSDs have provided failure rate statistics that give an indication of average reliability and lifespan for these drives.
Backblaze analyzed drive failure rates across over 41,000 consumer SSDs of varying makes/models deployed in their data centers over 6 years. They found failure rates varied significantly by manufacturer – between 1.5% to over 11% annually.
| Manufacturer | Annualized Failure Rate |
|---|---|
| Intel | 1.5% |
| Micron | 1.9% |
| Samsung | 2.2% |
| ADATA | 3.8% |
| Kingston | 4.7% |
| Crucial/Micron | 6.5% |
| Hynix | 6.7% |
| SanDisk | 7.5% |
| OCZ | 10.9% |
| Corsair | 11.6% |
Another analysis of SSDs used in data centers found the annual failure rate started around 1.5%, rising to 9% by the 5 year mark. Enterprise and server rated SSDs tend to have lower failure rates around 0.5 – 0.9% per year.
Overall SSDs still have a big edge over traditional HDDs in reliability. But their lifespans are still limited compared to the data retention ability of archived magnetic tapes which can last decades.
Conclusion
SSD failure is inevitable over time, but can also happen prematurely due to factors like excessive writes, overheating, physical damage or manufacturing flaws. Warning signs of a failing SSD include slower speeds, errors, bad sectors, unusual noises and problems booting. While data recovery from a dead SSD is costly and difficult, steps like monitoring health, preventing overuse, and taking backups can maximize an SSD’s lifespan. When failure does occur, replacement and safe disposal of the drive are necessary.