High Performance Computing (B.Sc.)

The seminar will take place during the following time slot: Mon, 12:15PM – 1:45PM. All of our meetings will be face-to-face in room 3073, CIP. Seminar participants must register in Friedolin. We have one presentation per week, so the number of participants is limited to 9.

Format

The seminar is divided into two parts. In the first part, we will read the book Computer Architecture: A Quantitative Approach (B1). The appendices A–C will be presented by the teaching staff; the chapters can be chosen by students. In the second part, we will read chapters 2 and 9–11 of the book Machine Learning Systems (B2).

The general format of the seminar is similar to a reading group. That is, all participants read the book chapter before attending the respective sessions. One person, either a student or teaching staff, becomes the expert on the topic. This person presents the topic for 20 minutes, followed by a short question round, and then leads the discussion.

Student Papers

All participants will write a scientific paper on their chosen seminar topic. The paper is due via email four weeks after the respective topic was discussed in the seminar. Use the ACM proceedings template with the sigconf option for your paper. The paper should be 4–6 pages in length (excluding references). You can write your paper in English or German.

Supervision

Preparing presentations and writing papers is hard. You can always ask for advice! Start early and keep in touch with your advisor!

Two meetings with your advisor are required:

  • The first meeting should be at least one week before your presentation.

  • The second meeting should be at least one week before your paper is due.

It’s your responsibility to schedule these meetings. The meetings are mandatory, otherwise you will not be admitted to the examination.

If illness interferes with your seminar presentation or paper, please let us know immediately. We will then find a solution.

Schedule

Date

What?

Chosen?

04/13

Kickoff

–

04/20

Instruction Set Principle (B1 Appx. A) and

Review of Memory Hierarchy (B1 Appx. B)

Felix

04/27

Pipelining: Basic and Intermediate Concepts (B1 Appx. C)

Tamino

05/04

Fundamentals of Quantitative Design and Analysis (B1 Ch. 1)

Marvin

05/11

Memory Hierarchy Design (B1 Ch. 2)

05/18

Instruction-Level Parallelism and Its Exploitation (B1 Ch. 3)

05/25

Whit Monday

–

06/01

Data-Level Parallelism in Vector, SIMD, and GPU Architectures (B1 Ch. 4)

06/08

Thread-Level Parallelism (B1 Ch. 5)

06/15

Warehouse-Scale Architectures for Utility Computing (B1 Ch. 6)

Elias

06/22

Domain-Specific Architectures (B1 Ch. 7)

06/29

ML Systems (B2 Ch. 2) and

Efficient AI (B2 Ch. 9)

Lukas

07/06

Model Optimizations (B2 Ch. 10) OR AI Acceleration (B2 Ch. 11)

Topics

Select one of the following chapters/chapter-pairs as your seminar topic. Topics will be assigned on a first-come, first-served basis.

  • Computer Architecture: A Quantitative Approach (book):

    • Fundamentals of Quantitative Design and Analysis (Ch. 1)

    • Memory Hierarchy Design (Ch. 2)

    • Instruction-Level Parallelism and Its Exploitation (Ch. 3)

    • Data-Level Parallelism in Vector, SIMD, and GPU Architectures (Ch. 4)

    • Thread-Level Parallelism (Ch. 5)

    • Warehouse-Scale Architectures for Utility Computing (Ch. 6)

    • Domain-Specific Architectures (Ch. 7)

  • Machine Learning Systems (book):

    • ML Systems (Ch. 2) and Efficient AI (Ch. 9)

    • Model Optimizations (Ch. 10) OR AI Acceleration (Ch. 11)

Generative AI

You are welcome to use generative AI tools when preparing your presentation and paper. However, you are fully responsible for the correctness of all content you submit or present! You must be able to explain and defend every claim and argument in your own words! Disclose any use of generative AI in the appendix of your paper.