course: Multi-Core Architectures and Parallel Programming Models

number:
141150
teaching methods:
lecture with tutorials
media:
computer based presentation
responsible person:
Prof. Dr.-Ing. Michael Hübner
Lecturers:
Prof. Dr.-Ing. Michael Hübner (ETIT), M. Sc. Jens Rettkowski (ETIT)
language:
german
HWS:
4
CP:
5
offered in:

goals

The students have an overview of multi-core architectures and parallel programming models. Using computer exercises the attendees can develop own embedded multi-core architectures based on FPGA technology and program state-of-the-art graphic cards using CUDA C/C++.

content

First multi-core architectures and their hardware components (e.g. processors, memories, and communication infrastructures) will be introduced. Afterwards parallel programming models (e.g. OpenMP, MPI, CUDA C/C++, and OpenCL) will be explained. The theoretical contents are supplemented using computer exercises for developing own multi-core architectures based on FPGA technology and for programming state-of-the-art graphic cards using CUDA C/C++.

requirements

keine

recommended knowledge

  • Knowledge of C/C++ or a similar programming language is required

miscellaneous

The lecture material will be distributed over Moodle.