The popularity of solid-state drives (SSDs) continues to grow, and this is no surprise. Compared to hard disk drives (HDDs), they are faster and smaller. They are also more energy-efficient and durable since they have no moving parts. Additionally, their capacity continues to increase while the cost keeps decreasing.
But are they really as reliable as they claim to be? Solid-state drives still have their flaws, and storage technologies that last for thousands of years are not commercially possible yet.
In this article, we’ll look into the reliability of SSDs. We will take a closer look at:
- Solid-state drive technology
- SSD storage memory
- Factors affecting reliability
- Signs of SSD failure
So, how reliable are SSDs? Let’s dive in.
SSD Technology
Almost all types of SSDs use NAND flash. Unlike many computer terms, NAND is not an acronym. Instead, it is a name derived from its logic gate, the basic building block of storage cells, called “NOT AND.” (Strangely, a NOT gate is a type of logic gate that only produces a false output when all its inputs are true.)
Flash memory (the term after NAND) refers to non-volatile solid-state storage. This means it can keep data even when the power is turned off.
NAND storage has specific characteristics that affect its lifespan. NAND flash works by storing data in individual memory cells organized in a grid array. When data is written to a NAND cell, it must be erased before new data can be written. Each time this happens, the insulator wears out as electrons are sent through it. Over time, this wear makes it harder for the insulator to hold the electrons in the right place. This difficulty can lead to errors in the stored data.
This means that the flash memory cells can only be reliably programmed and erased a finite number of times. This is measured in programming/erase cycles, more commonly known as P/E cycles.
P/E cycles are an important measure of SSD reliability. Nonetheless, there are other important factors to consider. These factors include TBW (Total Bytes Written) and MTBF (Mean Time Between Failures). Here are some definitions to help clarify these concepts:
Programmed/Erase Cycles
P/E cycles in solid-state storage involve writing data to NAND flash memory cells. Then, the data is erased to prepare for re-writing. The durability of an SSD is measured in P/E cycles. It varies depending on the technology. Usually, it ranges from 500 to 100,000 P/E cycles.
Terabytes Written (TBW)
TBW is the total amount of data that an SSD can handle. This is before it will fail. For example, here are the TBW ratings for the popular Samsung V-NAND SSD 870 EVO:
- 250GB model: 150TBW
- 500GB model: 300TBW
- 1TB model: 600TBW
- 2TB model: 1,200TBW
- 4TB model: 2,400TBW
All of these models have a 5-year warranty or until the TBW limit is reached, whichever comes first.
Mean Time Between Failures (MTBF)
MTBF is an indicator of the reliability of hardware components during their expected lifespan. For most components, MTBF is usually measured in thousands or even tens of thousands of hours. For example, the MTBF of an HDD is 300,000 hours, while an SSD be 1.5 million hours.
Manufacturers supply these specifications to help you understand the expected lifespan of the drive and its performance for specific applications.
Be careful when reviewing these specifications, as they do not guarantee that your specific SSD will last that long. Instead, they show expected error rates. These rates are based on a sample set of SSDs of the same model. A 1.2 million-hour MTBF suggests that a sample of 1,000 SSDs will face a failure every 150 days. This means about twice a year, assuming the drive is used eight hours a day on average.
Many SSDs today come with utilities to track the drive’s expected lifespan. These recommendations are based on monitoring the drive’s SMART attributes. As we discussed in earlier articles, different SSD manufacturers lack consistency. They vary in what attributes they check. They also differ in how they calculate the drive’s lifespan. If you want to use this information to decide when to replace your SSD, you should read the manual. It’s crucial to consult the manual for your specific SSD.
SSD Storage Memory
There are five types of NAND flash memory based on the number of bits each cell stores. We will discuss them below. Generally, as the number of bits per cell increases, the cost per bit decreases. Yet, durability and performance also decline.
SLC (Single Level Cell): One Bit Per Cell
SLC is the earliest type of NAND flash. Each cell stores one bit. SLC is fast and durable, but the space usage is low. This means that the physical size of SSDs using SLC is large.
MLC (Multi-Level Cell): Two Bits Per Cell
MLC stores two bits per cell. This roughly doubles the storage capacity and lowers the cost for a given form factor. Yet, MLC is slower because it has to differentiate between two bits in each cell.
TLC (Triple Level Cell): Three Bits Per Cell
TLC stores three bits per cell. This improves storage density and reduces cost, making SSDs more affordable. Nonetheless, TLC speeds are slower. Internal caching is introduced to improve performance. Read/write speeds are close to those of hard drives.
QLC (Quad Level Cell): Four Bits Per Cell
QLC is the current standard. Each cell stores four bits. This increases storage density, lowers price, and offers good speed with improved caching. But, the drive can wear out faster, especially when it is full.
3D NAND
In earlier technologies, storage cells were arranged in a single, 2D layer (called planar structure). In 3D NAND, cells are stacked in 3D. This increases storage density and speed but raises manufacturing costs and can reduce durability over time.
Factors to Consider for SSD Reliability
Generally, SLC and MLC are faster and more durable but have limited capacity. TLC and QLC store more data at a lower cost but are slower. Nonetheless, for average consumers, the speed difference may be insignificant and can sometimes be compensated with features like dynamic caching. 3D NAND is a good choice but comes at a higher price.
Compared to HDDs, SSDs are more durable. SSDs can better withstand accidental drops, shocks, and vibrations. They can also endure extreme temperatures and magnetic fields. This is because they don’t have moving parts like actuators and rotating disks. With their small size and low power consumption, switching from HDDs to SSDs is worth the time and effort.
But that’s not all. The performance and reliability standards should vary based on where the SSD is used. It is in a home desktop, data center, or Mars rover. SSD manufacturers are increasingly selling SSDs tailored to specific workloads (like write-intensive, read-intensive, or mixed-use). This means you can select the best level of SSD durability and capacity for your specific use case.
For example, enterprise users with high-volume databases choose drives that can handle more write cycles but sacrifice capacity. On the other hand, users of databases with infrequent writes select drives with higher capacity but lower performance. By doing this, manufacturers hide the complexities embedded in storage NAND (SLC, MLC, etc.), cache, and other technologies. This makes it easier to match your requirements with the best type of SSD.
Signs of SSD Failure
You have heard the scary clicking sound from a dying HDD. Since SSDs have no moving parts, you won’t hear an auditory warning before failure. Still, there are usually other signs that show it’s coming. If you notice any of these signs, take action promptly and replace the drive with a new one. Signs that your SSD is nearing the end of its life include:
- Errors Involving Bad Blocks
Just like bad sectors on an HDD, SSDs also have bad blocks. If you have a bad block, your computer will often try to read a file or save it. It will take an unusually long time and then fail. Your machine will eventually give up and show an error message. - Can’t Read or Write Files
Bad blocks affect files in two ways. First, the software detects a bad block while writing data to the drive, preventing the write. Second, it detects a bad block after data is written, preventing the read of that data. - File System Needs Repair
If you see an error message like this, the computer didn’t shut down properly. The computer was not shut down properly. The error also occurs due to bad blocks or other issues with the SSD. - Crash During Restart
A crash during a computer restart can show a problem with the drive. Make sure you have a current backup of all your data before it gets worse and the drive completely fails. - Drive Becomes Read-Only
Your drive may refuse to write more data and only allow reading. Fortunately, you can still retrieve data from the disk, and you should do so.
How reliable are SSDs?
Question 1: How long can we reasonably expect an SSD to last?
Answer: An SSD should last as long as the manufacturer expects, typically around five years, as long as it’s not used beyond its intended purpose (e.g., using QLC in write-heavy tasks). To make sure the SSD lasts, it’s a good idea to follow the manufacturer’s guidelines for proper use.
Question 2: Do SSDs fail faster than HDDs?
Answer: There are many variables when comparing the reliability of HDDs and SSDs, mainly how they are used. In the comparison of SSDs and HDDs, both work similarly in the first three years. Yet, after four years, HDDs have a higher failure rate. SSDs typically have an annual failure rate (AFR) of 1% or less in the first four years. Most SSD users replace their drives for reasons like upgrades for new technology. They also upgrade for larger capacity or faster speeds. This is rather than due to lifespan issues. Under normal use, SSDs should last for many years. If you upgrade your computer every three years, you won’t need to worry about the SSD’s lifespan. What matters most is that the SSD is reliable enough to avoid data loss during its use.
Question 3: Are SSDs suitable for long-term storage? Answer: Both SSDs and hard drives are designed for use. It’s not wise to put an external drive in a closet for years, whether it’s an SSD or an HDD. There is more anecdotal evidence about SSDs performing better than HDDs in this case. Nevertheless, it’s best to use external drives as part of a computer backup plan. Just don’t rely on them as the only backup.
Conclude
It’s important to know how different SSD technologies impact reliability. You should decide if it’s worth paying more for SLC over MLC or QLC. If your application doesn’t involve many write operations, choose a good quality SSD. Select one from a trusted brand. This choice will guarantee it lasts well.
Watch carefully for signs of failure or bad sectors. It is important regardless of the type of drive you are using. Make sure you have a reliable backup plan.