Jan Treibig

	Office:	Cauerstr. 6 Room 0.142
	E-mail:	Jan.Treibig@cs.fau.de
	Phone:	+49 9131 85-28677
	Fax:	+49 9131 85-28928
	Address:	Universität Erlangen-Nürnberg Lehrstuhl für Informatik 10 (Systemsimulation) Cauerstraße 6 D-91058 Erlangen

Research Interests

• Architecture specific optimization techniques for iterative solvers
• Parallel computing
• Object oriented programming techniques in simulation
• Computational fluid dynamics

I am currently working on the orcan and dime projects.

Publications

• Studienarbeit: Equivalent to bachelor thesis
  Numerische Simulation des Abklingverhaltens homogener isotroper Turbulenz, Erlangen 2000
  .ps (5.4 MB) , .ps.gz (1.2 MB)
• Diploma thesis
  Simulation von Gas-Feststoff-Mehrphasensystemen mit dem Lattice Boltzmann Verfahren, Erlangen 2002
  .ps (4.0 MB) , .ps.gz (1.8 MB) , BIBTEX
  
• Conference Papers:
  • G. Hager, T. Zeiser, J. Treibig, and G. Wellein: Optimizing performance on modern HPC systems: learning from simple kernel benchmarks, RGW 2005, .ps
  • J. Treibig, Simon Hausmann, Ulrich Ruede: Performance Analysis of the Lattice Boltzmann Method on x86-64 Architectures, ASIM 2005, .pdf
  • H. Hadler, J. Treibig, M. Kellner, T. Jung: Simulation Components in the Open Reflective Component Architecture, ASIM 2005, .pdf
  • J. Treibig, S. Bergler, U. Ruede: ORCAN: A platform for complex parallel simulation software, .pdf
• Other Publications (not fully reviewed):
  T. Zeiser, J. Treibig, G. Brenner, F. Durst: Simulation of highly loaded gas-solid two-phase flows by combining a cellular automata for the particles with a lattice Boltzmann flow solver, Int. J. Mod. Phys. B, 17 (2003) 201-204.
  

Talks

• Scalperf 2004, Bertinoro: Experiences with the design of cache-aware numerical algorithms (.ppt.gz)
• Scalperf 2005, Bertinoro: Optimizing Numerical Simulation Kernels for current Instruction and Memory Architectures (.pdf)
• ASIM 2005, Erlangen: Performance analysis of the Lattice Boltzmann Method on x86-64 Architectures(.pdf)
• Dagstuhl 2006, Schloss Dagstuhl: Why is the Performance Productivity Poor on Modern Computer Architectures?(.pdf)

Teaching

Master-/Batchelortheses (open and running)

• Diplomarbeit Markus Stuermer (2006): Optimierung des Mehrgitteralgorithmus auf IA64 Rechnerarchitekturen(pdf)
• Batchelorthesis Simon Hausmann (2005): Optimization and Performance Analysis of the Lattice Boltzmann Method on x86-64 based Architectures (pdf)
• Studienarbeit Markus Stuermer (2005): Optimierung des Red-Black-Gauss-Seidel-Verfahrens auf ausgewählten x86-Prozessoren (pdf - 7MB)
• Studienarbeit Silke Bergler (2005): Parallelisierung der Gitterkomponente im ORCAN Framework (ps - 20MB)
• Studienarbeit Florian Altendorfner (2003): Simulation des Einlaufbereichs eines Riserreaktors (pdf - 12MB)

Exercises

Have a look at the NuSiF Showroom.

• SS 2006: Exercises to NuSiF ( Numerical Simulation of Fluids)
• WS 2005/06: Architecture of High Performance Computers
• SS 2005: Exercises to NuSiF ( Numerical Simulation of Fluids)
• SS 2004: Exercises to NuSiF ( Numerical Simulation of Fluids)
• SS 2003: Exercises to NuSiF ( Numerical Simulation of Fluids)
• SS 2002: Exercises to NuSiF ( Numerical Simulation of Fluids), together with Marcus Mohr

Seminars

• WS 2005/06: Hardwarenahe Implementierung Performancekritischer Algorithmen

Courses

• SS 2006: C Course (CE Master): slides (pdf - 1MB)

PGP Key

This is my public key:

pub 1024D/CE7F540E 2004-02-09 Jan Treibig (work mail address) <jan.treibig@informatik.uni-erlangen.de>
Key fingerprint = 83B8 EB38 93DD 79B4 D381 D843 424E 16C7 CE7F 540E

You can download the key from here or from a public keyserver (e.g. http://www.dfn-pca.de/pgpkserv/).