In this course, you will learn to design the computer architecture of complex modern microprocessors. All the features of this course are available for free. It does not offer a certificate upon completion.
Motivation
Loading...
From the course by Princeton University
Computer Architecture
304 ratings
From the lesson
Introduction, Instruction Set Architecture, and Microcode
This lecture will give you a broad overview of the course, as well as the description of architecture, micro-architecture and instruction set architectures.
Meet the Instructors
David WentzlaffAssociate Professor
Electrical Engineering
So, what is computer architecture? Computer architecture, we're trying to
take an application, or something that a human wants to do, so for instance,
calculate a spreadsheet, play a video game, play some sort of music, and we want
to figure out how to map it down to physics.
So what do we mean by physics here? Well, physics is, we have in the world we
have particles and they bounce around, they interact, we have photons bouncing
around in the world, and we need somehow translate what humans want to do into what
the physics can do. And one of the problems with this is, this
is a really big gap. So how do you, go from application
directly to physics? You have physics bouncing around the
world. You have mechanical systems.
And without some level of abstraction, it's very hard to go from the application
directly to the physics. Now, in the broader sense, what computer
architecture is trying to do is it's trying to come up with the abstraction and
the implementation layers. That are needed to be able to bridge this
gap. So, we're gonna put a bunch of different
layers in here. And by making smaller abstraction layers
and smaller layers, we're going to be able to solve subsets of these problems and not
have to change everything here in the middle, just one.
Let's say the physics changes or the application changes a little bit, we'll be
able to reuse a lot of the work that we've done along the way.
And computer architecture is the study of these layers, and figuring out how to take
the physics and turn it into applications that humans want.
Now. I wanted to point out that in the natural
world, it's pretty challenging to directly jump from application to physics, but
there's a few examples. So one example I wanted to bring up
actually is the compass. The compass is a nifty little device that
directly takes physics and gets pushed up into an application.
But there's very few other examples. I think book is a okay example of this but
otherwise people build lots of abstraction layers in between here.
And one other thing I wanted to point out here is that we're trying to efficiently
use manufacturing technologies. So what I mean by that is, when physics
changes, or we move to smaller transistors, or we move from, let's say,
silicon to gallium arsenide or some other implementation technology, that changes
the bottom layer here, but we want to still be able to reuse a lot of the other
work that we've done over the years. So, let's take a look at the abstractions
in modern day computing systems. We start off here, physics.
We have fundamental physical laws of how particles interact, and we move up to
devices. So, what do I mean by devices?
We have transistors, and we can build different types of transistors.
Mosfets, BJTs, we can build other, other types of FETs, and we start to build
circuits, bigger circuits out of this, and out of those we go and build gates.
And from gates we can go to RTL, or registered transfer language here, which
is sort of our verilog coding. And then we start to get into what this
course is about, which is different types of architecture.
So there's microarchitecture, which is how you do you go and actually build a
specific implementation of a chip. And then we're, above that we have
instruction set architecture, which gives us some portability on top of that.
And then we get into operating systems and virtual machines, Programming languages,
and algorithms. And then finally we get to the algorithm
the application programmer sitting on top. So in this course, in computer
architecture, we're only gonna be focusing in this three middle layers here,
instruction set architecture, or what I will sometimes refer to as big A computer
architecture. Microarchitecture, or, sometimes what
people call organization, of computing systems.
And, register transfer language and we'll, we'll also overlap a little bit into the,
the abstraction layers above and below here.
So, we'll talk a little bit about some operating system concerns and virtual
machine concerns. And we'll talk a little bit about how the
tran-, the technology and the gates influences the computing system.
So one important point here about computer architecture is it's constantly changing
because different constraints and different applications are changing.
So, people have new applications that they come up with.
They come up with ways, or people come up with different applications.
I now want to have a smart phone. Well, that didn't exist twenty years ago.
And this pushes down different requirements and these different
requirements actually suggest how to change the architecture.
If you, for instance want to do a lot of video processing that can actually
influence your computer architecture so you add specialized instructions to do
video processing. Likewise technology constraints push up,
so as we go to smaller and smaller transistors, we'll say the smaller and
smaller transistors let's say go faster, but the wires go slower, well that's going
to influence your computer architecture. And a lot times new, new technologies make
new architecture possible. So, what do I mean by that?
Well, lets say all of sudden, you get a big bump in transistors, you get twice as
many transistors. Well now, architectures and micro
architectures that didn't use to make sense, start making sense, because you can
for instance fit a lot more computation on one chip.
And, what's, the interesting thing here, is that, computer architecture is not done
in a vacuum. So, computer architecture actually
provides feedback up and down this abstraction layer stack.
So, it will give feedback, and it will actually influence the research directions
that technology looks at. And will influence research directions and
influence different applications that are possible.
So this is not all done in a vacuum, the computer architect actually sits in a very
key location in this abstraction layer stack here, cuz you can push up and push
down, and you're not just forced to work with what you're given, if you will, from
a technology perspective. But that might take a few years.
So let's talk a little bit more about what this class is about, but we'll, we'll do
it by way of a little bit of history. And to put a little bit of a Princeton
connection in here, we're gonna talk computers back in the late '40s, early
'50's, nineteen, 1940s, 1950s. So, here's actually a, a picture of the
IAS, or the Institute for Advanced Study Machine, which was built in the Institute
of Advanced Study, which is maybe about a mile and a half away from this classroom.
And it was designed by Jon Von Goyman, and, and to give a little bit of insight
here. It was first moved in Princeton in 1952,
it was actually started in the late'40's and took them a couple years to get, to
get working. And one of the interesting things here is
that this machine is actually built out of vacuum tubes.
So, everyone thinks about transistors today.
Well, before we had transistors we had little glass tubes that actually looked
like light bulbs, and inside of those there were switches that could be
switched. So, very similar idea to what our
transistor can do, but, instead you had a evacuated glass tube, and you had a little
transmitter, it was a cathode ray tube. And then you had a gate that could open
and close inside of this. So people were building computers long
before transistors, and people were thinking about computer architecture long
before transistors. And people were even thinking about
computer architecture and these sorts of technologies even before vacuum tubes.
So there was electromechanical systems, a good example of this actually was
originally in phone systems. They had these cool electromechanical
switches. So when you take your old rotary phone and
you sort of turn it and then it goes, tick, tick, tick, tick, tick, tick, tick,
tick, tick. What that's actually doing is, it's
sending pulses which are turning a mechanical, a little mechanical arm inside
of a relay system, and people built computers out of those electro-mechanical
relays. So you can have switches that turn, change
switches, similar sorts of ideas to transistors.
And even before that people had looked at building mechanical systems.
So there's mechanical adding machines for instance.
And now of days we, of course, have transistors.
So, it was computing then, in the fifties and if we fast forward to today, we have
lot, computers look very different. In this figure here, ya know, this thing
was the size of a room. Pretty big room TV scale sort of, a person
is maybe yea tall in this figure thing. It's a, sort of normal-sized room, but,
still sort of room sized. But today we have lots of different
applications. And computing looks very different.
So, we have computing, let's say, in small systems.
So we have a little sensor network-, networks, and little sensor nodes
distributed. We have fancy cameras.
We have smartphones. We have mobile audio players and iPods.
We have laptops, like my laptop here. We have self-driving cars, we have big
servers, we have Xboxes, and there's a lot of variety now in what computing systems
look like. So, the influence of this technology,
computer architecture, has been very, very broad.
And we even go on to seeing things like routers, flying, unmanned autonomous
vehicles. We have GPS's, which are little computers
that can basically fit on your wrist these days and tell you exactly where you are.
E books, tablets, set-top boxes and the list goes on and on and on.
And what, what, I want to get across here is that computer architecture has a very
rich history. And this history is continuing and it's
very relevant today. So we're not studying something which no
one cares about anymore. People are sitting there sort of ready to
get the next generation computer architecture.
People, and, and it used to be you know, you want your faster desktop computer.
And that may not be as important today, but what is important is people want their
faster smartphone. They want to enable voice recognition on
the go. They want to be able to, in scientific
applications, they want to build a model, some health system that's really complex,
that you weren't able to do before. So, it continues on and on.
It is very relevant today, and it has a very rich history.
In this course we're gonna talk a little bit about the history, we'll mostly focus
on the technology. Sometimes when people teach computer
architecture classes, they have much more emphasis on the history.
This class, we're gonna Touch on history a little bit, but more focused on the, on
the technology considerations. So here's a chart that's from Hennessy and
Patterson's Computer Architecture, or A Quantitative Approach, and what this graph
is trying to show is, is something very fundamental to computer architecture, and
it's what's been driving our industry. So what we see here is we see different
processor designs plotted on a log plot. So this is ten, 100, 1000. This is a log
plot and it says performance versus years. So, if you look at this, this plot, you'll
see that, well this roughly looks like a straight line, and a straight line on a
log plot is an exponential increase in performance.
So, we've seen computing going up exponentially faster.
And this is a, a really fundamental to driving what's been going on in our
industry and why computer architecture is so important.
And I do want to say that, you know, this exponential increase is, comes from two
things. It's not all computer architects, I'd love
to able to sit, stand here and say, "we did all this." Well, no, a fair amount of
this is from getting, better, better technology.
What I mean that is the, lower down layers and the implementation technology, like
the transistor technologies. And some of it is from having better and
better computer architecture. And what is really important here to note
is even if you have better and better transistors.
A lot of times what happens if you look at this graph is what's happening is you're
getting more transistors, but those transistors are not necessarily
exponentially faster. So, what computer architects have to do is
we need to figure out how to take buckets and buckets of more transistors and turn
them into more performance. And that's what this is many times called
is it's called Moore's Law. If you've heard that term before, what it
is, is we're trying, Moore's, Gordon Moore said that every eighteen months to two
years, you're going to get twice as many transistors which, you can have for the
same amount of dollars. That was what was originally said.
People sort of transformed that now into meaning your computer's gonna get twice as
fast every year. That's not what, Gordon Moore originally
actually said. He said he can get twice as many
transistors for a certain amount of dollars.
And people have also sometimes taken this to mean that you get twice as many
transistors on a chip every year. It's not quite what he said, but close
enough approximation. And when, I don't quite have the graph
here. But if you look at computing in general
across, this, this is all across transistor based technologies.
But if you go farther back into the past, you can actually plot other technologies,
like vacuum tube technologies, and relay based technologies.
And it also fits on this, this graph relatively well.
So sort of, if you continue down here, you're gonna see vacuum tubes show up.
And it's sort of still on this exponentially increasing curve.
Okay, so let's look at, there's two inflection points in this graph that we
wanna look at. First one's right here, you can see that
the slope changes a little bit. Well, what, what happened here?
This was the introduction of reduced instruction set computers, or risk
computers. So we got a little bit of a, a crank up
there when people came out with the first risk.
And, another thing you notice is, this graph keels over a little bit here, and,
we are gonna be talking a lot about this in this course, is, what happens or why,
did this happen? So, what, what happened here?
Well, sequential performance, so this is the performance of a single program; was
getting faster exponentially. But then, I don't know, depending on who
you ask. I like to use 2005 as the number but
somewhere between 2003 and 2007, sequential processor performance started
to, to really have a, a problem. But overall performance of your processor,
still continuous to go up today. And what happened is we had to move to
multi-core processors or multiple cores on a single chip.
And hopefully, the hope is, that this graph will continue on here with multiple
cores, If we can figure out how we can effectively paralyze our programs, versus
our sequential performance tapering, tapering off.
Cuz it would be very harmful to computer architecture and computing industry if all
of a sudden our computers stopped getting faster, no one would be buying new
computer chips.
Explore our Catalog
Join for free and get personalized recommendations, updates and offers.
Coursera provides universal access to the world’s best education,
partnering with top universities and organizations to offer courses online.
© 2018 Coursera Inc. All rights reserved.
