Skip to content
Hi, Bot

Hard Disk Drive

Spinning mirrors and a flying needle: the cheapest way to keep a lot of data.

How big?About the size of a paperback; the platters spin 120 times a second.

Real: What it looks like.

Keys: arrows rotate · + / − zoom · 0 reset · 1–4 views · S signal · T tour · L labels · Space spin. Models are stylised and built from code: proportions are honest, details are simplified.

What it is

Data is stored as tiny magnetic patches on glass or aluminium platters. A read/write head flies a few nanometres above the surface on a cushion of air (or helium), closer than a fingerprint is thick.

To read something, the arm swings to the right track and waits for the spot to spin underneath. That wait is why hard drives are slow at random access but fine for long, steady reads.

Why AI needs it

Data lakes hold petabytes of raw text, images and video, and keeping all of it on flash would cost a fortune. Big object-storage clusters are mostly hard drives, with copies spread so a dead drive loses nothing.

Every labeled part

  1. 1

    Platters

    Mirror-smooth discs coated in magnetic material. Signal mode spins them.

  2. 2

    Spindle motor

    Spins the platters, typically 7,200 times a minute.

  3. 3

    Read/write heads

    One per platter surface, flying nanometres above it.

  4. 4

    Actuator arm

    Swings the heads across the tracks in a few thousandths of a second.

  5. 5

    Voice-coil magnet

    A strong magnet and coil move the arm, the same principle as a loudspeaker.

  6. 6

    Controller board

    Underneath. Turns magnetic wiggles into bits.

  7. 7

    Data + power

    SATA or SAS connection to the server.

Try it · concept lab

Spinning vs. solid state

Compare random reads with one long read.

Hard drive

≈ 100–200 reads/s

The arm has to swing and wait for the spot to spin round, every time.

SSD

≈ 500,000+ reads/s

No moving parts: any cell, any time, many at once.

Animation is slowed way down. Typical figures; drives vary.

Big idea: Moving parts make random access slow. Hard drives are for big, steady, cheap storage; SSDs are for fast access.

Try it · concept lab

Break some drives

Tap drives to kill them. Which files survive?

Tap drives to break them.

File A: readableFile B: readableFile C: readableFile D: readable

Each file lives on 3 drives. Any 2 can die.

Big idea: At data-lake scale drives fail every day. Spreading copies, or clever math pieces, means nothing is lost.

Swap it: other ways to do the same job

  • SSD

    Hundreds of times faster at random reads, no moving parts, still several times the cost per terabyte.

  • Tape

    Even cheaper and lasts decades on a shelf, but you wait minutes to fetch something.

  • Cloud object storage

    Rent it instead of buying it. Underneath, it's racks of these drives.

Talk about it

  1. Q1

    A hard drive is a spinning disk with a tiny arm floating above it. What other spinning things store sound or pictures?

  2. Q2

    Big storage systems keep extra copies so one broken drive loses nothing. How many copies of your favorite photo would you keep?

For grown-ups: there are no right answers here. Ask a question, then ask "why do you think that?" The reasons matter more than the answer.

Printable question sheet (PDF)