High-Performance Computing for Cyber-Physical Systems

ED160021International Experience3 ECTSEnglishsummer semesterLehrstuhl für Cyber-Physical Systems in Production Engineering (Prof. Caccamo)

AI-edited module sheet. Based on the TUMonline module description, edited for readability.Original in TUMonline

What it is about

You will learn advanced techniques of High-Performance Computing (HPC) with Python to develop efficient and scalable programs for Cyber-Physical Systems (e.g., transportation networks, energy systems, manufacturing). By the end you will be able to parallelize Python code for multicore and multiprocessor systems, design efficient database access, and apply HPC methods to embedded multi-core boards as well as accelerators.

What you will be able to do

Understand theoretical concepts of High-Performance Computing
Write and optimize performant Python code
Use multiprocessing to utilize multiple processes
Use shared memory and synchronization mechanisms
Apply multi-threading for I/O-bound tasks
Design efficient access to large databases
Design and implement HPC solutions for various CPS platforms
Consider sustainability aspects in energy-efficient implementation

What the module consists of

LectureConveying fundamental concepts and methods of software design for CPS through presentations and demonstrations
lab-based tutorialsProject work (1–2 submissions) for practical application of lecture content with direct supervisor feedback

Teaching method

LectureExplanation of concepts and methods as well as limited demonstrations
Tutorial/Project workPromotes active learning through practical implementation and direct feedback on developing efficient software
Discussions and practical activitiesEnable design, implementation and evaluation of HPC solutions for CPS

What you should already know

Good understanding of computer science and technology
Prior knowledge of Python (recommended)recommended
Knowledge of another programming language (recommended)recommended

Assessment

Written examLaboratory projectCoursework

The module examination consists of a written final exam (60 minutes, no aids, examination contribution 100% of the exam) and an exercise component in the form of laboratory projects (coursework with 1–2 submissions, pass/fail). In the laboratory projects you will demonstrate the practical application of multiprocessing, shared memory, multi-threading for I/O, and efficient database access. The exam will include comprehension and programming tasks to verify fundamental concepts.

Reading list

The Python Language Reference - https://docs.python.org/3/reference/index.html
Micha Gorelick, Ian Ozsvald - High Performance Python - Second Edition
Steven F. Lott, Ricardo Bánffy - Functional Python Programming - Third Edition
Michal Jaworski, Tarek Ziadé - Expert Python Programming - Fourth Edition

Dates

LectureHigh-Performance Computing for Cyber-Physical Systems

Wed10:00–11:30
2104, Seminar (5501.02.104)
14× · 14.10.–03.02.
- 14.10.
- 21.10.
- 04.11.
- 11.11.
- 18.11.
- 25.11.
- 02.12.
- 09.12.
- 16.12.
- 23.12.
- 13.01.
- 20.01.
- 27.01.
- 03.02.

ExerciseTutorial High-Performance Computing for Cyber-Physical Systems

Mon10:00–11:30
2104, Seminar (5501.02.104)
7× · 26.10.–01.02.
- 26.10.
- 09.11.
- 23.11.
- 07.12.
- 21.12.
- 18.01.
- 01.02.

From the current semester, not binding.

Show TUMonline data

Sprache: Englisch
Turnus: Sommersemester
Modulniveau: Bachelor/Master
Moduldauer: Einsemestrig
Gesamtstunden: 90
Präsenzstunden: 45
Eigenstudiumstunden: 45
Organisationsname: Lehrstuhl für Cyber-Physical Systems in Production Engineering (Prof. Caccamo)

Courses

High-Performance Computing for Cyber-Physical Systems
High-Performance Computing for Cyber-Physical Systems
High-Performance Computing for Cyber-Physical Systems - Übung

Official page in TUMonline · Details are not binding.