1 - Rechnerarchitektur [ID:10869]
50 von 2363 angezeigt

Okay, let's start. Welcome to the first lecture on computer science, semester 17 and 18.

I don't think it's easy to read. I think we need some darkening here, right?

Over there.

Can you see it?

On the board.

Which one?

There's one more.

All might.

Okay, so we can get some cinema atmosphere here.

So, very briefly, I think we had a basic computer lecture before.

It's nice that you came here, not us.

It's not that bad either, right?

Normal average rate of failure. I always wonder why.

30% is high, but well, there will be some reasons.

Those who don't know me yet, my name is Dietmar Fei.

I'm the head of computer architecture in the lecture hall.

And as a result, I also hold the deepening lecture in computer architecture.

As I said, I actually had a lot of space. So I got it assigned.

Or rather, they gave me a very small room. Okay, I'm the one to blame, I admit it.

Because in StudOn there was something in it, maximum number of participants, 15.

That was sometime before 2010, when I started here.

Then I got a small room in the new building.

I had to stack everything. Then we immediately looked for the one next to it.

That worked out last year, too, from Mathematics.

We immediately gave it to the university.

About a quarter of an hour later I got an e-mail from the space assignment from Mathematics.

What a stupidity! So, not like that.

Okay, then I went back to the DUC position and looked for another one.

I actually had a nice room in the new building. Wonderful, I was satisfied.

Then the second round of this planning tool came and threw me out.

There is now a concept mod exercise in it and they gave me this stack room again.

I panicked, looking for where I could find another one.

And which one was still there? The most beautiful room in the entire university, this one.

Good, but it's not so nice outside in winter either.

So we'll do it in the winter semester here, pretty cozy.

Okay, let's start. Today I will bring some things to the organizational front.

What is it about? So, pre-reading, computer reading.

17, 18, wonderful. I even adjusted that properly yesterday.

What I haven't done is change the batteries.

Good, can you read now? It works like that, okay.

Do you know this one? Who is that guy?

Do you know anyone? A certain Johann von Neumann, John von Neumann.

He is standing here in front of the IAS computer, Institute of Advanced Studies in Princeton.

So, yes, this is a computer that was built according to his ideas.

But I'll send it away right away. We are all nationalists abroad, I can tell you.

But at one point I would like to say, who invented the computer?

Not him, or Zuse, he was on it before.

He may have lived in the wrong country at the wrong time,

so in the time of the National Socialist Regime, but he was on it before.

The Americans recognize that now, they don't say it that loudly, but okay.

Teil einer Videoserie :

Zugänglich über

Offener Zugang

Dauer

01:33:14 Min

Aufnahmedatum

2017-10-16

Hochgeladen am

2019-04-30 14:49:03

Sprache

de-DE

Die Vorlesung baut auf die in den Grundlagen der Rechnerarchitektur und -organisation vermittelten Inhalte auf und setzt diese mit weiterführenden Themen fort. Es werden zunächst grundlegende fortgeschrittene Techniken bei Pipelineverarbeitung und Cachezugriffen in modernen Prozessoren und Parallelrechnern behandelt. Ferner wird die Architektur von Spezialprozessoren, z.B. DSPs und Embedded Prozessoren behandelt. Es wird aufgezeigt, wie diese Techniken in konkreten Architekturen (Intel Nehalem, GPGPU, Cell BE, TMS320 DSP, Embedded Prozessor ZPU) verwendet werden. Zur Vorlesung werden eine Tafel- und eine Rechnerübung angeboten, durch deren erfolgreiche Beteiligung abgestuft mit der Vorlesung 5 bzw. 7,5 ECTS erworben werden können. In den Tafelübungen werden die in der Vorlesung vermittelten Techniken durch zu lösende Aufgaben vertieft. In der Rechnerübung soll u.a. ein einfacher Vielkern-Prozessor auf Basis des ZPU-Prozessors mit Simulationswerkzeugen aufgebaut werden. Im Einzelnen werden folgende Themen behandelt:
  • Organisationsaspekte von CISC und RISC-Prozessoren

  • Behandlung von Hazards in Pipelines

  • Fortgeschrittene Techniken der dynamischen Sprungvorhersage

  • Fortgeschritten Cachetechniken, Cache-Kohärenz

  • Ausnutzen von Cacheeffekten

  • Architekturen von Digitalen Signalprozessoren

  • Architekturen homogener und heterogener Multikern-Prozessoren (Intel Corei7, Nvidia GPUs, Cell BE)

  • Architektur von Parallelrechnern (Clusterrechner, Superrechner)

  • Effiziente Hardware-nahe Programmierung von Mulitkern-Prozessoren (OpenMP, SSE, CUDA, OpenCL)

  • Leistungsmodellierung und -analyse von Multikern-Prozessoren (Roofline-Modell)

Empfohlene Literatur
  • Patterson/Hennessy: Computer Organization und Design
  • Hennessy/Patterson: Computer Architecture - A Quantitative Approach

  • Stallings: Computer Organization and Architecture

  • Märtin: Rechnerarchitekturen

Einbetten
Wordpress FAU Plugin
iFrame
Teilen