‘They run forever’ and other myths about SSDs
Myth - Solid state devices and solid state storage have no moving parts and therefore fail very rarely.
Reality – Solid state storage is far more reliable than traditional HDD, as they have no moving parts, but they are most commonly installed in mobile devices, which are subjected to much harsher environmental conditions than most traditional HDD applications. The most common cause of solid state device failure Kroll Ontrack sees is by water. This causes electrical short circuits and damage to the electronic components. In comparison, less than 1% of traditional HDD failures are due to water damage.
Myth - You don’t need a backup if you sync to the cloud from your SSD.
Reality – There is an old computing rule: “Unless you backup to two different locations you don’t have a valid backup”. This rule does not change because the particular storage media you are using is more reliable than others. Kroll Ontrack often helps customers recover their data after both their local and cloud versions of a file become inaccessible, are corrupted, deleted or overwritten. What happens is the local version gets damaged and then syncs to the cloud. Most Cloud providers save previous versions of files but this does not always work. We always recommend a second backup is made.
Myth - SSDs can be easily repaired if they go wrong.
Reality – Wrong, the difficulty with SSD recoveries is based on the way that data is stored on the device. An SSD controller (marked on the picture) scrambles and sometimes encrypts the data when it gets written to the memory chips. This creates a unique pattern/map of the data and the way it is stored on the chips. Additional difficulty is created when manufacturers start to use different controllers as each controller creates its own pattern/map of the data. The only way to recreate it is to reverse engineer this pattern manually and extract the data.
Myth - Shutting down the computer incorrectly or running the battery until it is flat does not harm the data.
Reality – Wrong, removing the power while the SSD is reading or writing data requires the SSD to shut down into a safe state within milliseconds not to lose any data. This is a tall order and to expect it to work every time without fail is hoping the controller designers got their sums right. From Kroll Ontrack engineers’ experience we know occasionally, just occasionally, they get it wrong. We know this because we see the results - data corruption has occurred and the user has lost access to their files.
Myth - SSDs will run forever as they don’t wear out.
Reality - Wrong, the life expectancy and capacity of all media reduces over time. A design phenomenon known as write amplification causes the capacity and the speed of the SSD to reduce overtime. This is due to the fundamental way the SSD erases data from memory.
The useful life expectancy of a SSD could be as little as five years under heavy use. It is envisaged that most people upgrade their hardware every three years so this should not be a problem unless you buy second-hand. There are currently two types of memory used in SSDs: SLC (single-layer-cell), and MLC (multiple-layer cell). SLC lasts much longer but costs more to manufacture. All the cheaper drives and most USB flash memory drives are MLC based devices. Flash memory can only be programmed and erased a limited number of times. This is often referred to as the maximum number of program/erase cycles (P/E cycles) it can sustain over the life of the flash memory. Single-level cell (SLC) flash, designed for higher performance and longer endurance, can typically operate between 50,000 and 100,000 cycles. As of 2011, multi-level cell (MLC) flash is designed for lower cost applications and has a greatly reduced cycle count of typically between 3,000 and 5,000.
Myth - Using encryption has no significant disadvantages.
Reality - Wrong, encryption significantly reduces the recoverability of data if a HDD is accidentally reformatted. Using full disk encryption compounds this problem. Many data loss situations caused by logical software problems cannot be recovered from. If the failure causes damage to the encryption disk and this prevents the disk from mounting, this in-turn means the data cannot be decrypted. It also means for badly damaged disks a partial or targeted recovery cannot be attempted as the location of the important data cannot be identified. It is an all or nothing recovery or it would be like doing a 10,000 piece jigsaw where all the pieces are the same colour and the same shape!