Flash memory has largely revolutionized how we deal with computer storage, and it has enabled both faster and smaller computers that are safer from data loss thanks to the fact that they don’t require any moving parts. You may not know, however, that there’s more than one kind of flash memory — and while they can be similar, there are some key differences.
Before we dive into the different types of NAND storage, its important to understand what NAND, or flash storage actually is. NAND is essentially a non-volatile storage medium that doesn’t need power in order to retain data — unlike some other storage mediums. NAND can, however, exist in multiple different kinds.
But what are those different kinds? And why are some better than others? Here’s an overview of each of the major flash memory types and why they’re different.
SLC storage, AKA single-level cell storage, is the most common type of flash storage — and the fastest. To understand single-level cell storage, you first need to understand how data storage works.
In general terms, flash storage works through transistors existing in one of two states — which represents either a 1 or a 0. When many of those transistors, or cells, which store what are called bits, are strung together, they store data. That’s what data is — strings of bits, each of which is either a 1 or a 0.
Because of the fact that each cell only stores one bit, data can be accessed much faster than other types of flash storage — however the trade-off is that SLC storage normally offers lower data capacities. Single-level cell storage also has the highest cost.
Single-level cell storage is often used in high-performance memory cards and other mission-critical situations.
Image courtesy of Intel Free Press | Flickr
MLC, or multi-level cell, is a type of memory element in which more than one bit of information can be stored in each cell. In other words, each cell has multiple levels, which means that more bits can be stored with the same number of transistors.
So why is that different from other types of storage? Well, in single-level cell NAND flash technology, a transistor can exist in one of two states — which equates to either a 1 or a 0, meaning that each transistor represents one bit.
Of course, there is a trade-off — and that’s memory speed. The main benefit of MLC technology is that it offers a lower cost per unit of storage, which in turn leads to a higher density of data for the same price.
There’s a secondary type of MLC storage, called eMLC, or enterprise multi-level cell. This type of storage has been enhanced for more write cycles than traditional, consumer-grade MLC flash storage. Consumer-grade MLC storage generally only offers between 3,000 and 10,000 write cycles, while eMLC cells can offer up to 20,000 or even 30,000 write cycles. eMLC generally has a longer life because of an advanced controller operating the cell.
Single-level and multi-level cells aren’t the only type of flash storage. Perhaps a better name for “multi-level” cell storage would be “dual-level cell,” as triple-level cell storage actually has its own name.
As the name suggests triple-level cell storage stores a hefty three bits of information per cell. This technology was first developed by Samsung, and Samsung actually refers to it as 3-bit MLC.
Everything that’s bad about MLC storage, however, is amplified with TLC storage — that’s to say that TLC storage is even lower cost, but its even slower and less reliable.
There’s a trend here — the more levels, the less costly — but also, the more levels the slower and less reliable a storage medium is. Single-level cell storage is by far the best performing flash storage type, but it may not be the best for all situations — sometimes, getting more storage that might be slightly lower performing is simply necessary.