Flash memory is everywhere. It exists in your USB memory stick, your camera’s SD card, your solid state hard drive, hospital medical equipment, industrial robotics machines, and countless more devices and gadgets.
But have you ever stopped to wonder what it actually is? Whether there are different types of flash memory? What they are used for? How does it all work?
In this article, I’ll explain some of the most common types of flash memory. You’ll be an expert in no time!
Flash Memory Explained
There are a few different types of flash memory, but NAND is arguably the most common. It’s the one you’ll find in USB cards, MP3 players, and other devices that require high-capacity data storage.
All flash memory has two key characteristics:
- Non-volatile — Non-volatile memory does not need a power supply to retain its data. As such, it’s most commonly used for long-term storage that persists between reboots. An example of its opposite (AKA volatile memory) is your computer’s RAM. RAM loses all retained information when you shut down your PC.
- Finite Number of Write Cycles — Because of the way it works, flash memory can be used a limited number of times before it starts to wear out. Individual cells will slowly fail and performance will degrade.
How Does It Work?
Flash memory stores data in an array of cells and each cell holds at least one bit of data. The cells are organized into blocks, where a block is defined as a contiguous set of bytes that form an identifiable unit of data.
A block is the smallest programmable/erasable portion of the array. The blocks are written to by electrical charge, with each cell either representing a 1 or a 0 number.
When all of the blocks are considered together, they form a memory chip. The chip is mounted on a printed circuit board, which also includes a basic controller and a USB interface.
NAND itself is raw flash memory and uses its own protocol. Designs which implement NAND — such as SD cards and solid state drives — often add microcontrollers on top to implement a Flash Translation Layer (FTL). The FTL translates your disk usage (for example, via USB) into meaningful NAND operations.
The Different Types of NAND
NAND flash memory is a catch-all phrase. There are many different designs and sub-classes. The three most common ones are the only ones you really need to worry about.
SLC (Single Level Cell)
SLC is widely considered to be the best version of NAND. It stores one bit of data per memory cell and therefore has the best endurance, being able to handle approximately 100,000 write cycles per cell before deterioration.
It has the fastest write speeds and the lowest power consumption, but can be up to three times more expensive than a basic Three Level Cell design and often has lower storage capacity. It’s best deployed in high-performance, medium-density situations.
It is typically used in situations where speed is required — for example, servers, high-performance media cards, hybrid disk drives, and top-end solid state drives. It can also be found in high-end professional SD cards, such as Panasonic’s FX Series.
MLC (Multi Level Cell)
MLC NAND stores two bits per cell and can, therefore, hold twice the amount of data in the same sized device, significantly reducing the cost-per-bit. It’s great for higher-density, low-cycle applications.
However, there is a trade-off: MLC can only support around 10,000 write cycles per cell before deterioration, thus negatively impacting its longevity.
MLC can be considered consumer-grade NAND. It’s responsible for nearly 80 percent of all NAND flash shipments worldwide and is used most commonly in consumer-class solid state drives.
TLC (Three Level Cell)
TLC NAND is the cheapest of the three forms, costing about 30 percent cheaper than MLC memory (and even cheaper than SLC memory).
It’s the highest density — able to save three bits of data per cell — but also has the worst durability. In fact, a typical TLC chip can only support about 4,000 write cycles per cell, which is far worse than both MLC and SLC.
TLC is most commonly used in cost-efficient products that don’t require top-end NAND performance and won’t last nearly as long as its counterparts. Examples include MP3 players, USB memory sticks, and low-end portable media devices.
It will not be found on anything that runs an operating system or stores critical data (such as solid state drives).
What About eMMC?
eMMC stands for “Embedded Multimedia Card”, which itself grew out of its predecessor, MMC (Multimedia Card).
Multimedia cards first hit the shelves back in 1997. They were used as a storage medium for portable devices, including the earliest MP3 players and digital cameras. Ports for the cards were often built into computers at the time, but as the popularity of SD cards grew, fewer manufacturers bothered with MMC anymore.
Today, you’ll find it difficult to find a PC with an MMC slot.
However, the legacy has lived on in the form of eMMC cards. eMMC memory is still widely used in the mobile sector as the most common form of integrated storage in mobile devices, and can even be found in some low-end PCs, tablets, and Chromebooks.
It’s arranged on a small ball grid array (BGA) that’s soldered onto the device and non-removable. It’s slower and therefore cheaper than other forms of NAND, so if you can afford it, prefer aiming for NAND devices instead.
NAND and eMMC in a Nutshell
I hope I’ve explained NAND and eMMC in an understandable way, but if you’re still confused, don’t worry. It’s undoubtedly a confusing topic.
The confusion is compounded by the speed at which technology moves. For example, the latest version of the eMMC now boasts write speeds rivaling discrete SATA-based SSDs at around 400 MB/s. But for the most part, you only need to know the differences between SLC, MLC, and TLC.
If you’re confused about some of the terminology or which type of NAND is right for you, get in touch below. I’ll do my best to help you out.