Contents:
Facing increasing power dissipation and little instruction-level parallelism left to exploit, computer architects are realizing further performance gains by placing multiple "slower" processor cores on a chip rather than by building faster uniprocessors. As a consequence, application performance will increasingly depend on the efficient use of parallelism.
Students attending this seminar will be assigned topics related to state-of-the-art technology in the field of parallel programming. Based on the assignment, every student will perform the following tasks under the guidance of a supervisor:
• Search for relevant literature
• Write a topic summary
• Prepare a talk of 40 min during which the topic is presented to other participants
Instructors:
- Prof. Dr. Felix Wolf
- Dr. Daniel Becker
- Alexandru Calotoiu
- Marc-André Hermanns
- Monika Lücke
- Suraj Prabhakaran
- Sebastian Rinke
- Aamer Shah
- Christian Siebert
- Zoltan Szebenyi
Topics:
A list of topics can be found
here.
Schedule:
The schedule can be found
here.
Topic summary deadline: 22.12.2011
The talk is due one week prior the date when it is to be presented.
Requirements:
We recommend but do not demand to have attended the following events:
• Course Parallel Programming I
• Literature research training offered by Renate Eschenbach-Thomas from the Computer Science Library
Certificates:
Certificates will be issued based on written reports, oral presentations, and active participation. The certificates can be graded if required.
Language:
Presentations and reports can be either in English or in German.