What is NAND?
In PC hardware, NAND almost always means NAND flash memory: the non-volatile memory chips inside SSDs, USB flash drives, SD and microSD cards, and the built-in storage of phones and tablets. Non-volatile means it keeps data with the power off. An SSD is essentially a set of NAND chips plus a controller that decides where each block of data goes.
How NAND flash stores data
Each memory cell is a transistor that traps electrical charge, and the amount of charge represents the stored bits. Cells are written in pages and erased in much larger blocks, and a page can't be overwritten in place: the old data must be erased first. Every program/erase cycle wears the cell a little, so NAND has a limited write life.
Storing more bits per cell raises capacity and lowers cost, but the charge levels sit closer together, so endurance and write speed drop:
| Type | Bits per cell | Trade-off |
|---|---|---|
| SLC (single-level cell) | 1 | Highest endurance and speed, lowest density, most expensive |
| MLC (multi-level cell) | 2 | High endurance, more density than SLC |
| TLC (triple-level cell) | 3 | A balance of cost and endurance; widely used in consumer SSDs |
| QLC (quad-level cell) | 4 | Highest density and lowest cost per GB, lowest endurance |
3D NAND stacks layers of cells vertically to fit more capacity in each chip.
How SSDs manage wear
- Wear leveling: the controller spreads writes across all blocks so no area wears out early.
- TRIM and garbage collection: the OS tells the SSD which blocks hold deleted data, so the controller can erase them in the background and keep writes fast.
- Spare blocks and error correction: extra blocks replace worn ones, and ECC fixes small bit errors.
- SLC caching: many TLC and QLC drives use part of their flash as a fast SLC cache, so very large transfers slow down once it fills.
Endurance is rated in TBW (terabytes written) or drive writes per day. S.M.A.R.T. values such as percentage used or wear leveling count show how much life is left.
NAND vs. NOR, and the NAND gate
NOR flash is the other main flash type. It allows fast random reads of individual bytes but is less dense, so it's used for small firmware chips, such as the BIOS/UEFI chip on many motherboards.
Both are named after logic gates. A NAND (NOT AND) gate outputs 0 only when all of its inputs are 1; if any input is 0, the output is 1. NAND flash got the name because the cells in each string are wired in series, like the transistors in a NAND gate.
For the CompTIA A+ exam
NAND isn't on the Core 1 (220-1201) acronym list, but it's inside the SSDs, flash drives and memory cards covered by objective 3.4, and objective 5.2 covers drive troubleshooting, including S.M.A.R.T. failures. Know that SSDs use NAND flash with no moving parts, that flash wears out with writes, and that TRIM and wear leveling help it last.