623.714 (18W) Selected Topics in Distributed Multimedia Systems: Parallel Computing

Wintersemester 2018/19

Anmeldefrist abgelaufen.

Erster Termin der LV
08.11.2018 14:00 - 16:00 , S.2.69
... keine weiteren Termine bekannt

Überblick

Lehrende/r
LV-Titel englisch
Selected Topics in Distributed Multimedia Systems: Parallel Computing
LV-Art
Vorlesung-Kurs (prüfungsimmanente LV )
Semesterstunde/n
2.0
ECTS-Anrechungspunkte
4.0
Anmeldungen
10 (25 max.)
Organisationseinheit
Unterrichtssprache
Englisch
mögliche Sprache/n der Leistungserbringung
Englisch
LV-Beginn
08.11.2018

LV-Beschreibung

Intendierte Lernergebnisse

The objective of this course is to provide an extensive overview of the fi eld of parallel computing, with emphasis on the use of

OpenMP and MPI to develop parallel programs for multicores, and an introduction to GPU programming using CUDA and

OpenCL. 

Lehrmethodik

At the end of this course, participants should be able to:

• Understand the fundamentals of parallel computing

• Analyze a problem and identify the adequate parallelization model

• Write message-passing and shared memory programs

• Design parallel solutions for new problems

• Use computational models to estimate applications performance

Inhalt/e

Parallel computing is also called the third pillar of science as many fields from medicine, weather forecast, automobile industry, among many others, are using computing simulation for new developments. Additionally, parallel computing is becoming the current computing paradigm as hardware tends to multi-processing units. The common desktop and laptop are today built with a multicore processor that collectively has more processing power, than a single core processor, but cores are individually less powerful. Accelerators, such as GPUs, are also becoming a commodity, which allows the common user to have access to high–performance machines. Programmers will have to deal with multiprocessor architectures in order to use effectively the machines of today and of the future.

There are several approaches to program a multicore machine, being the shared memory model the straight step to implement a parallel version of a given sequential code. The well known library pthreads allows the implementation of this model but with a level of detail that is not always desirable. To overcome this burden, OpenMP emerged as a reliable and efficient library to develop shared memory programs on multicore processors. For GPUs there are special languages, such as CUDA and OpenCL, that follow also a shared memory approach.

 The distributed memory model is another approach to obtain a parallel version of a sequential code, and is suitable when more than one processor cooperates to run a given application. MPI is the most popular library to implement this model, which is based on the exchange of messages among processors.

Prüfungsinformationen

Prüfungsmethode/n

- Project 60% (groups of 2 students)
- Written evaluation (4 questions): 40%

Prüfungsinhalt/e

Will be announced soon!

Beurteilungskriterien/-maßstäbe

- Project 60% (groups of 2 students)
- Written evaluation (4 questions): 40%

Beurteilungsschema

Note/Grade Benotungsschema

Position im Curriculum

  • Masterstudium Angewandte Informatik (SKZ: 911, Version: 13W.1)
    • Fach: Distributed Multimedia Systems (Wahlfach)
      • Selected Topics in Distributed Multimedia Systems ( 2.0h VK / 4.0 ECTS)
        • 623.714 Selected Topics in Distributed Multimedia Systems: Parallel Computing (2.0h VC / 4.0 ECTS)
  • Masterstudium Game Studies and Engineering (SKZ: 992, Version: 17W.2)
    • Fach: Gebundenes Wahlfach (Wahlfach)
      • Modul: Game Engineering
        • 4.1 Fundamental Topics I in Distributed Multimedia Systems ( 0.0h VC / 4.0 ECTS)
          • 623.714 Selected Topics in Distributed Multimedia Systems: Parallel Computing (2.0h VC / 4.0 ECTS)
            Absolvierung im 1., 2., 3. Semester empfohlen
  • Masterstudium Game Studies and Engineering (SKZ: 992, Version: 17W.2)
    • Fach: Gebundenes Wahlfach (Wahlfach)
      • Modul: Game Engineering
        • 4.1 Selected Topics in Distributed Multimedia Systems ( 0.0h VC / 4.0 ECTS)
          • 623.714 Selected Topics in Distributed Multimedia Systems: Parallel Computing (2.0h VC / 4.0 ECTS)
            Absolvierung im 1., 2., 3. Semester empfohlen

Gleichwertige Lehrveranstaltungen im Sinne der Prüfungsantrittszählung

Diese Lehrveranstaltung ist keiner Kette zugeordnet