An increasing number of cores in embedded systems implies great challenges for industry and science ensuring software reliability. We design and develop an HPC software platform for the verification of complex reliability requirements for embedded systems. On the one hand we exploit high performance computing in the context of verification problems of the automotive industry, on the other hand, we explore the efficient use of current and future HPC architectures with new approaches. The project has four main objectives: - Highly scalable software verification and analysis of concurrency scenarios for embedded applications in the automotive industry, - development of new scaling methods for integer dominated applications relying on irregular communication patterns, - gathering experience on optimizations for manycore and exascale applications class, concurrency simulation and software verification, - transfer of the developed tools, methods and processes to generic HPC consulting services for other applications.

Publications

2019
Learned Clause Minimization in Parallel SAT Solvers Pragmatics of SAT 2019, pp. 1-11, 2019 Marc Hartung, Florian Schintke BibTeX
arXiv
HPSV
Pinpoint Data Races via Testing and Classification 2019 IEEE International Symposium on Software Reliability Engineering Workshops (ISSREW); 3rd International Workshop on Software Faults (IWSF 2019), pp. 386-393, 2019 Marc Hartung, Florian Schintke, Thorsten Schütt BibTeX
DOI
HPSV
ZIB_Glucose: Luby Blocked Restarts and Dynamic Vivification Proceedings of SAT Race 2019 : Solver and Benchmark Descriptions, pp. 43-44, 2019 Marc Hartung PDF
BibTeX
HPSV