Course of study, specific field and term:

• Certificate in Artificial Intelligence (compulsory), Artificial Intelligence, 1st semester
• Master Entrepreneurship in Digital Technologies 2020 (advanced module), technology field computer science, arbitrary semester
• Master Computer Science 2019 (basic module), technical computer science, 1st or 2nd semester
• Master Medical Informatics 2019 (optional subject), technical computer science, 1st or 2nd semester
• Master Robotics and Autonomous Systems 2019 (optional subject), Elective, 1st or 2nd semester
• Master IT-Security 2019 (basic module), technical computer science, 1st or 2nd semester
• Master Medical Informatics 2014 (basic module), computer science, 1st or 2nd semester
• Master Entrepreneurship in Digital Technologies 2014 (basic module), technology field computer science, 1st or 2nd semester
• Master Computer Science 2014 (basic module), technical computer science, 1st or 2nd semester

Classes and lectures:

• Parallel Computer Systems (lecture, 2 SWS)
• Parallel Computer Systems (exercise, 2 SWS)

Moodle Page of the Course

Workload:

• 100 Hours private studies
• 60 Hours in-classroom work
• 20 Hours exam preparation

Contents of teaching:

• Motivation and limitations for parallel processing
• Parallel computing models
• Taxonomy of parallel computers
• Multi/manycore-systems
• Graphic Processing Units (GPUs)
• OpenCL
• Specification languages
• Hardware architectures
• System management of many-core systems

Qualification-goals/Competencies:

• Students are able to characterize different parallel computing architectures.
• They are able to explain models of parallel computing.
• They are able to make use of common programming interfaces for parallel computing systems.
• They are able to judge which kind of parallel computing system is best suited for a dedicated problem and how many cores should be used.
• They are able to evaluate the pros and cons of different hardware architectures.
• They are able to write programs for parallel computing systems under considerations of the underlying hardware architecture.
• They are able to compare methods for dynamic voltage and frequency scaling (DVFS) for manycore systems.

Grading through:

• written exam

Responsible for this module:

• Prof. Dr.-Ing. Mladen Berekovic

Teacher:

• Institute of Computer Engineering
• Prof. Dr.-Ing. Mladen Berekovic

Literature:

• G. Bengel, C. Baun, M. Kunze, K. U. Stucky: Masterkurs Parallele und Verteilte Systeme - Vieweg + Teubner, 2008
• M. Dubois, M. Annavaram, P. Stenström: Parallel Computer Organization and Design - University Press 2012
• B. R. Gaster, L. Howes, D. R. Kaeli, P. Mistry, D. Schaa: Heterogeneous Computing with OpenCL - Elsevier/Morgan Kaufman 2013
• B. Wilkinson; M. Allen: Parallel Programming - Englewood Cliffs: Pearson 2005
• J. Jeffers, J. Reinders: Intel Xeon Phi Coprozessor High-Performance Programming - Elsevier/Morgan Kaufman 2013
• D. A. Patterson, J. L. Hennessy: Computer Organization and Design - Morgan Kaufmann, 2013

Language:

• offered only in German

Notes

Prerequisites for attending the module:
- None

Prerequisites for the exam:
- Successful completion of homework assignments during the semester.

Last updated:

9.9.2020