Name: 內存與存儲。計算機科學速成班 #19 (Memory & Storage: Crash Course Computer Science #19)
Uploaded: 2021-01-14T07:55:46.000Z
Duration: 12 min 17 s
Description: 【看影片學英語】數萬部 YouTube 影片，搭配英漢字典即點即查，輕鬆掌握單字發音與用法，長久累積看電影不必再看字幕。

Hi, I'm Carrie Anne, and welcome to Crash
Course Computer Science!

程度副詞

We've talked about computer memory several
times in this series, and we even designed

In general, computer memory is non-permanent.

If your xbox accidently gets unplugged and
turns off, any data saved in memory is lost.

For this reason, it's called volatile memory.

What we haven't talked so much about this
series is storage, which is a tad different.

Any data written to storage, like your hard
drive, will stay there until it's over-written

It used to be that volatile memory was fast
and non-volatile storage was slow, but as

computing technologies have improved, this
distinction is becoming less true, and the

Nowadays, we take for granted technologies
like this little USB stick, which offers gigabytes

of memory, reliable over long periods of time, all at low cost, but this wasn't always true.

The earliest computer storage was paper punch cards, and its close cousin, punched paper tape.

By the 1940s, punch cards had largely standardized
into a grid of 80 columns and 12 rows, allowing

for a maximum of 960 bits of data to be stored
on a single card.

The largest program ever punched onto cards,
that we know of, was the US Military's Semi-Automatic

Ground Environment, or SAGE, an Air Defense
System that became operational in 1958.

The main program was stored on 62,500 punchcards,
roughly equivalent to 5 megabytes of data,

that's the size of an average smartphone
photo today.

Punch cards were a useful and popular form
of storage for decades, they didn't need

power, plus paper was cheap and reasonably
durable.

However, punchcards were slow and write-once,
you can't easily un-punch a hole.

So they were a less useful form of memory,
where a value might only be needed for a fraction

of a second during a program's execution,
and then discarded.

A faster, larger and more flexible form of
computer memory was needed.

An early and practical approach was developed
by J. Presper Eckert, as he was finishing

His invention was called Delay Line Memory,
and it worked like this.

You take a tube and fill it with a liquid,
like mercury.

Then, you put a speaker at one end and microphone
at the other.

When you pulse the speaker, it creates a pressure
wave.

This takes time to propagate to the other
end of the tube, where it hits the microphone,

converting it back into an electrical signal.

And we can use this propagation delay to store
data!

Imagine that the presence of a pressure wave
is a 1 and the absence of a pressure wave

Our speaker can output a binary sequence like
1010 0111.

The corresponding waves will travel down the
tube, in order, and a little while later,

hit the microphone, which converts the signal
back into 1's and 0's.

If we create a circuit that connects the microphone
to the speaker, plus a little amplifier to

compensate for any loss, we can create a loop
that stores data.

The signal traveling along the wire is near
instantaneous, so there's only ever one

bit of data showing at any moment in time.

But in the tube, you can store many bits!

After working on ENIAC, Eckert and his colleague
John Mauchly, set out to build a bigger and

better computer called EDVAC, incorporating
Delay Line Memory.

In total, the computer had 128 Delay Lines,
each capable of storing 352 bits.

That's a grand total of 45 thousands bits
of memory, not too shabby for 1949!

This allowed EDVAC to be one of the very earliest
Stored-Program Computers, which we talked

However, a big drawback with delay line memory
is that you could only read one bit of data

If you wanted to access a specific bit, like
bit 112, you'd have to wait for it to come

around in the loop, what's called sequential
or cyclic-access memory, whereas we really

want random access memory, where we can access
any bit at any time.

It also proved challenging to increase the
density of the memory, packing waves closer

together meant they were more easily mixed
up.

In response, new forms of delay line memory
were invented, such as magnetostrictive delay

These delay lines use a metal wire that could
be twisted, creating little torsional waves

By forming the wire into a coil, you could
store around 1000 bits in a 1 foot by 1 foot square.

However, delay line memory was largely obsolete
by the mid 1950s, surpassed in performance,

reliability and cost by a new kid on the block:
magnetic core memory which was constructed

out of little magnetic donuts, called cores.

If you loop a wire around this core…. and
run an electrical current through the wire,

we can magnetize the core in a certain direction.

If we turn the current off, the core will
stay magnetized.

If we pass current through the wire in the
opposite direction, the magnetization direction,

1 bit of memory isn't very useful, so these
little donuts were arranged into grids.

There were wires for selecting the right row
and column, and a wire that ran through every

core, which could be used to read or write
a bit.

In each of these little yellow squares, there
are 32 rows and 32 columns of tiny cores,

So, each of these yellow squares could hold
1024 bits.

In total, there are 9 of these, so this memory
board could hold a maximum of 9216 bits, which

The first big use of core memory was MIT's
Whirlwind 1 computer, in 1953, which used

And, instead of just a single plane of cores,
like this, it was 16 boards deep, providing

Importantly, unlike delay line memory, any
bit could be accessed at any time.

This was a killer feature, and magnetic core
memory became the predominant Random Access

Memory technology for two decades, beginning
in the mid 1950s even though it was typically

Although starting at roughly 1 dollar per
bit, the cost fell to around 1 cent per bit

Unfortunately, even 1 cent per bit isn't
cheap enough for storage.

As previously mentioned, an average smartphone
photo is around 5 megabytes in size, that's

Would you pay 4 hundred thousand dollars to
store a photo on core memory?

If you have that kind of money to drop, did
you know that Crash Course is on Patreon?

字幕列表影片播放

內存與存儲。計算機科學速成班 #19 (Memory & Storage: Crash Course Computer Science #19)

episode

stick

approach

access