Programming Techniques for Supercomputers (Lecture and Tutorial) (PTfS)7.5 ECTS
(englische Bezeichnung: Programming Techniques for Supercomputers (Lecture and Tutorial))
(Prüfungsordnungsmodul: Wahlpflichtbereich Informatik)

Lehrveranstaltungen:

• • Programming Techniques for Supercomputers
    (Vorlesung, 4 SWS, Gerhard Wellein, Mo, Mi, 16:15 - 17:45, H10; Do, 8:15 - 9:45, H10; First lecture: April 23rd, 2019 (Wednesday). There will be one lecture on Monday and the other one Wednesday OR Thursday (final decision after first lecture))
  • Programming Techniques for Supercomputers
    (Übung, 2 SWS, Gerhard Wellein, Mo, 14:15 - 15:45, 00.153-113; Mi, 10:15 - 11:45, 0.01-142; First tutorials: Monday, April 29th)

Empfohlene Voraussetzungen:

Experience in C/C++ or Fortran programming required; basic knowledge of MPI and OpenMP programming

Inhalt:

• Introduction to the architecture of modern supercomputers
  

• Single core architecture and optimisation strategies
  

• Memory hierarchy and data access optimization
  

• Concepts of parallel computers and parallel computing
  

• Efficient “shared memory” parallelisation (OpenMP)
  

• Parallelisation approaches for multi-core processors including GPUs
  

• Efficient “distributed memory” parallelisation (MPI)
  

• Roofline performance model
  

• Serial and parallel performance modelling
  

• Energy efficient implementation and execution of parallel programs

Lernziele und Kompetenzen:

Students

• acquire a comprehensive overview of programming modern supercomputers efficiently for numerical simulations
  

• learn modern optimisation and parallelisation strategies, guided by structured performance modelling
  

• acquire an insight into innovative programming techniques and alternative supercomputer architectures
  

• are able to implement basic numerical methods with high hardware efficiency on parallel computers
  

• are able to program and use modern supercomputer with high (energy) efficiency
  

Literatur:

G. Hager and G. Wellein: Introduction to High Performance Computing for Scientists and Engineers. CRC Computational Science Series, 2010. ISBN 978-1439811924
J. Hennessy and D. Patterson: Computer Architecture. A Quantitative Approach. Morgan Kaufmann Publishers, Elsevier, 2003. ISBN 1-55860-724-2

Weitere Informationen:

Verwendbarkeit des Moduls / Einpassung in den Musterstudienplan:

1. Computational Engineering (Rechnergestütztes Ingenieurwesen) (Master of Science): ab 1. Semester
   (Po-Vers. 2013 | TechFak | Computational Engineering (Rechnergestütztes Ingenieurwesen) (Master of Science) | Wahlpflichtbereich Informatik | Wahlpflichtbereich Informatik)

Studien-/Prüfungsleistungen:

Programming Techniques for Supercomputers (Lecture and Tutorial)_ (Prüfungsnummer: 278169)

Prüfungsleistung, Klausur, Dauer (in Minuten): 90, benotet, 7.5 ECTS

Anteil an der Berechnung der Modulnote: 100.0 %

weitere Erläuterungen:
Written exam will also cover material presented and exercised in the tutorials. In der Klausur wird auch Stoff der Übungen abgeprüft.

Prüfungssprache: Englisch

Erstablegung: SS 2019, 1. Wdh.: WS 2019/2020, 2. Wdh.: keine Wiederholung

1. Prüfer:

Gerhard Wellein