§44.1–44.3Flash: Storing Bits, and the Basic Operations

Solid-state drives (SSDs) solid-state drive An SSD: persistent storage built from transistors (flash), with no moving parts — unlike a hard disk. It presents the standard block read/write interface but internally uses a flash translation layer to hide flash's quirks (erase-before-write, wear-out). Much faster than disks at random I/O, but pricier per GB. defined in ch. 44 — open in glossary store data persistently with no moving parts — just transistors, like memory and CPUs. The technology is NAND flash flash memory NAND-based flash: solid-state storage that traps charge in transistors to store bits and retains it without power. Its defining quirks: you can read/program at page granularity but must ERASE a whole (larger) block before re-programming any page in it, and blocks wear out after many program/erase cycles. defined in ch. 44 — open in glossary , and it has two quirks that make building an SSD a real challenge: to rewrite a small chunk you must first erase a bigger chunk (expensive), and writing a spot too many times wears it out.

44.1 Storing a Single Bit

A flash cell stores bits by trapping charge at distinguishable levels — a multi-level cell multi-level cell How many bits a flash transistor stores by distinguishing charge levels: SLC (single-level, 1 bit — fastest, longest-lived, priciest), MLC (2 bits), TLC (3 bits — densest/cheapest, slower, shorter-lived). More bits per cell = more capacity per dollar but worse performance and endurance. defined in ch. 44 — open in glossary design trades density against speed and endurance:

44.2 Blocks and Pages

Flash cells are grouped into banks, and a bank is accessed at two granularities:

• A flash block flash block The unit of flash ERASE (an 'erase block'), typically 128 KB–2 MB — much larger than a flash page, and different from a disk/FS block. Erasing sets every bit to 1 and destroys all pages in the block, so any data you still need must be copied out first. This erase-before-program constraint shapes all of SSD design. defined in ch. 44 — open in glossary (or erase block, 128 KB–2 MB) — the unit of erase.
• A flash page flash page The unit of flash reads and programs (writes), a few KB (e.g. 4 KB) — distinct from a VM page or a disk block. Pages live within a flash block. A page can be programmed only once after its block is erased; changing it again requires erasing the whole block. defined in ch. 44 — open in glossary (a few KB) — the unit of read and program. Each block holds many pages.

Here’s a bank as three blocks of four pages each:

44.3 The Basic Operations

Flash supports three low-level operations:

• Read a page: fast (~10s of µs), uniform regardless of location — flash is a random-access device (unlike a seeking disk).
• Erase a block: ~milliseconds, and it resets every bit in the block to 1, destroying all its pages’ contents. Required before any re-programming.
• Program a page: ~100s of µs, flips selected 1s to 0s to write the page’s data. Only works on an erased page.

So each page moves through states — INVALID → (erase) → ERASED → (program) → VALID — and once VALID, the only way to change it is to erase its whole block:

Why writes are expensive

That erase-before-program rule is the whole story of SSD design. Overwriting a single page forces erasing its entire block — which wipes every other page in that block too:

Reading is easy and fast (great random-read performance, far beyond a disk). Writing is the hard part — erase the block (after saving any data you need), then program — and repeating this program/erase cycle is what eventually wears the flash out. The rest of the chapter is about taming these costs.