This title appears in the Scientific Report : 2015 

10,000 Performance Models per Minute – Scalability of the UG4 Simulation Framework
Vogel, Andreas (Corresponding author)
Calotoiu, Alexandru / Strube, Alexandre / Reiter, Sebastian / Nägel, Arne / Wolf, Felix / Wittum, Gabriel
Jülich Supercomputing Center; JSC
Euro-Par 2015: Parallel Processing / Träff, Jesper Larsson (Editor) ; 2015, Chapter 40 ;
Berlin, Heidelberg Springe 2015
519 - 531
978-3-662-48096-0 (electronic)
978-3-662-48095-3 (print)
10.1007/978-3-662-48096-0_40
21st International Conference on Parallel and Distributed Computing, Vienna (Austria), 2015-08-24 - 2015-08-28
Contribution to a conference proceedings
Computational Science and Mathematical Methods
Lecture Notes in Computer Science 9233
Please use the identifier: http://dx.doi.org/10.1007/978-3-662-48096-0_40 in citations.
Numerically addressing scientific questions such as simulating drug diffusion through the human stratum corneum is a challenging task requiring complex codes and plenty of computational resources. The UG4 framework is used for such simulations, and though empirical tests have shown good scalability so far, its sheer size precludes analytical modeling of the entire code. We have developed a process which combines the power of our automated performance modeling method and the workflow manager JUBE to create insightful models for entire UG4 simulations. Examining three typical use cases, we identified and resolved a previously unknown latent scalability bottleneck. In collaboration with the code developers, we validated the performance expectations in each of the use cases, creating over 10,000 models in less than a minute, a feat previously impossible without our automation techniques.