VVZ API is not affiliated with ETH Zurich. Data might be outdated or incorrect. Please view the official ETHZ Vorlesungsverzeichnis for binding information.

151-0116-00L 7 Credits BSC D-MATH , D-INFK

High Performance Computing for Science and Engineering (HPCSE) for CSE

Lecturers: Dr. Sergio Miguel Martin, Dr. Eva Athena Economides
Examiners: Dr. Jens Honoré Walther, Dr. Sergio Miguel Martin, Dr. Eva Athena Economides
VVZ CR n/a

Last Updated: 2026-02-05 16:22:56

Abstract

This course focuses on programming methods and tools for modern parallel systems, such as large-scale supercomputers with multi and many-core processors. Emphasis will be placed on techniques and models to maximize the performance of such systems. This is a hands-on course that relies on practical applications in science and engineering to demonstrate the importance of HPC.

Objective

The objective of this course is to specialize students in the use of supercomputer systems and advanced (GPU) processors for solving large-scale scientific and engineering applications. Students will acquire tools that will enable them to solve computational problems, both in scientific research and engineering, that require large amounts of computation which can only be achieved by the efficient use of supercomputers and GPU processors.

Content

The topics offered by this lecture include: - Large-scale computing topics: communication-tolerant programming and scalability. + Communication-Tolerant Programming + Hybrid Parallelism (MPI + OpenMP) + Work Tiling and Advanced Threading-Based Libraries - High-Throughput Computing and it's use in Monte-carlo and sampling methods for stochastic optimization methods and uncertainty quantification (UQ) - Principles and advance performance optimization topics for Many-Core (GPU) Programming

Resources

Lecture Notes

https://www.cse-lab.ethz.ch/teaching/hpcse-ii_fs23/The materials include class notes, presentation slides, and lecture recordings.

Literature

- Class notes - Introduction to High Performance Computing for Scientists and Engineers, G. Hager and G. Wellein - CUDA by example, J. Sanders and E. Kandrot

Learning Materials (Links)

General Information

Language
English
Levels
BSC
Frequency
Yearly recurring

Examination

Type
session examination
Mode
written 180 minutes
Aids
You are allowed to bring a HANDWRITTEN summary of 7 A4 sheets, written on the front and back pages (14 pages total). Photocopies are not allowed.
Digital
The exam takes place on devices provided by ETH Zurich.
This course will be offered for the last time in Spring Semester 2023.The class has one compulsory continuous performance assessment (mandatory project, comprising of 6 biweekly assignments).The final grade will be determined as a weighted average of the grades: 70% session examination and 30% project.The project will be divided into 6 homework assignments, each counting to 5% of the course grade, delivered and graded every 2 weeks.All assignments must be delivered on the due date. Late assignments will be awarded a grade of 1.The assignments rely on each other so it would be more difficult to do only few than all of them. The assignments are envisioned as critical elements of the class and as assistance to the successful completion of the exam. The exam will contain a written part and exercises on the computer and it will contain material that refers directly to the assignments in the project.

Course Components

Type Title Time & Place Hours
lecture with exercise High Performance Computing for Science and Engineering (HPCSE) II
This course will be offered for the last time in Spring Semester 2023. Lecture: 14-16h Exercises: 10-12h. The exercises begin in the second week of the semester.
  • Mon 10:15-12:00 (ML H 44)
  • Mon 14:15-16:00 (ML H 44)
4 h weekly
practical/laboratory course High Performance Computing for Science and Engineering (HPCSE) for CSE
This course will be offered for the last time in Spring Semester 2023.
  • Fri 08:15-10:00 (HG E 26.1)
2 h weekly

Offered In