ENHANCE aims on a better integration and simplified usage of heterogeneous computing resources within current and upcoming computing systems.

Heterogeneous systems contain multiple compute components like multi-core processors, complemented by graphics processing units (GPUs) and/or field programmable gate arrays (FPGAs). Employing such hardware architectures raises several challenges in programmability, performance estimation and scheduling that are approached within the ENHANCE project and shall result in a framework enabling the development and use of applications on heterogeneous systems. The benefit of the developed methods for the industrial application partners is of special importance.

Publications

2014
An Automated Approach for Estimating the Memory Footprint of Non-linear Data Objects Euro-Par 2013: Parallel Processing Workshops, pp. 249-258, Vol.8374, Lecture Notes in Computer Science, 2014 (preprint available as ) Sebastian Dreßler, Thomas Steinke BibTeX
DOI
ENHANCE
2013
Automated Analysis of Complex Data Objects 28th International Supercomputing Conference, ISC 2013, Leipzig, Germany, June 16-20, 2013, 2013 Sebastian Dreßler, Thomas Steinke BibTeX
ENHANCE
Tool-Supported Integration of Hardware Acceleration in Automotive CFD-Simulations NAFEMS 2013 Proceedings, 2013 F. Ries, M. Ditze, A. Piater, E. Singer, V. Fäßler, Sebastian Dreßler, T. Soddemann BibTeX
ENHANCE
2012
A Novel Hybrid Approach to Automatically Determine Kernel Interface Data Volumes ZIB-Report 12-23 Sebastian Dreßler, Thomas Steinke PDF
BibTeX
URN
ENHANCE
Energy consumption of CUDA kernels with varying thread topology Computer Science - Research and Development, pp. 1-9, 2012 Sebastian Dreßler, Thomas Steinke BibTeX
DOI
ENHANCE
On Improving the Performance of Multi-threaded CUDA Applications with Concurrent Kernel Execution by Kernel Reordering Application Accelerators in High Performance Computing (SAAHPC), 2012 Symposium on, pp. 74-83, 2012 Florian Wende, Frank Cordes, Thomas Steinke BibTeX
DOI
ENHANCE