Course

PAP - Parallel Programming


PDF Course Catalog Deutsche Version: PAP

Version: 2 | Last Change: 01.10.2019 15:54 | Draft: 0 | Status: vom verantwortlichen Dozent freigegeben

Long name Parallel Programming
Approving CModule PAP_MaMT, PAP_MaTIN
Responsible
Prof. Dr.-Ing. Arnulph Fuhrmann
Professor Fakultät IME
Level Master
Semester in the year summer semester
Duration Semester
Hours in self-study 78
ECTS 5
Professors
Prof. Dr.-Ing. Arnulph Fuhrmann
Professor Fakultät IME
Requirements The exercises require programming knowledge and the use of console-oriented programs in Linux-based operating systems.
Language German, English if necessary
Separate final exam Yes
Literature
P. Pacheco: An Introduction to Parallel Programming, Morgan Kaufmann, 2011
T. Rauber, G. Rünger: Parallele Programmierung, Springer, 2012
T. Rauber, G. Rünger: Multicore: Parallele Programmierung, Springer, 2007
R. Oechsle: Parallele und verteilte Anwendungen in Java, Hanser, 2011
B. Goetz, J. Bloch, J. Bowbeer, D. Lea, D. Holmes, T. Peierls: Java Concurrency in Practice, Addison-Wesley Longmanm 2006
Jason Sanders: CUDA by Example: An Introduction to General-Purpose GPU Programming, Addison-Wesley Longman, 2010
Aaftab Munshi: OpenCL Programming Guide, Addison-Wesley Longman, 2011
Final exam
Details
In a final examination (written, optional oral), the students demonstrate their knowledge and competences summarily. The examination includes exemplary parts of the course.
Minimum standard
Achieving the individual minimum score per exam, typically 50% of the maximum score.
Exam Type
In a final examination (written, optional oral), the students demonstrate their knowledge and competences summarily. The examination includes exemplary parts of the course.

Learning goals

Knowledge
- Basic concepts, models and technologies of parallel processing (parallelism, concurrency, SISD, SIMD, MISD, MIMD, loose- and closely coupled systems, distributed systems)
- Parallel performance measures (speedup, efficiency)
- Architecture of GPUs
- Parallel Algorithms for GPUs
Expenditure classroom teaching
Type Attendance (h/Wk.)
Lecture 2
Tutorial (voluntary) 0
Special literature
keine/none
Special requirements
none
Accompanying material
Lecture notes (slides)
Separate exam
none

Learning goals

Skills
- Analyze and structure tasks related to programming parallel programs, assign relevant parallel hardware architecture and transfer to parallel design
- Implement parallel programs (multicore hardware with threads and GPUs)
- Analyze parallel programs using suitable tools and present results in a comprehensible way
- Estimate and analyze performance of parallel programs
- Derive information from original English sources and standards
Expenditure classroom teaching
Type Attendance (h/Wk.)
Practical training 2
Tutorial (voluntary) 0
Special literature
keine/none
Special requirements
none
Accompanying material
Exercises, server systems, GPU systems
Separate exam
Exam Type
solving exercises within limited functional / methodical scope
Details
The principles, models, methods, technologies and tools conveyed in the lecture will be deepened and practiced in the practical course on the basis of current tasks in the context of media-based and/or interactive systems. The students work independently on the exercises.
Minimum standard
80% of the exercise tasks has been successfully completed.

© 2022 Technische Hochschule Köln