PD Dr. Matthias Korch

PD Dr. Matthias Korch

Contact

Email: matthias.korch@uni-bayreuth.de
Phone: +49 (0) 921/55-7705
Fax: +49 (0) 921/55-7702
Room: 2.11
Building: Angew-Inf

Teaching

  • Computer Architecture and Computer Networks (WS 2008/2009 to WS 2013/2014)
  • Parallel Algorithms (since SS 2011)
  • Parallel and Distributed Systems I (since WS 2014/2015)
  • Programming Innovative Computer Architectures (since SS 2014)
  • Seminars for Bachelor and Master students on different topics of parallel computing, e.g., multi-core processors, load balancing and scheduling, processor organization and computer architecture, programming GPUs and hardware accelerators, high performance computing, memory hierarchies
  • Introduction to Programming for Teacher Candidates  (SS 2009, SS 2010)
  • Formal Languages and Compiler Construction (SS 2008)
  • Practical courses and projects for Master students
  • Bachelor and Master Theses

research topics

  • Parallel computing
  • Systems of ordinary differential equations
  • Irregular algorithms
  • Distributed systems

Projects

Active Projects

Finished Projects

Curriculum vitae

Further information

Cooperations

Program Committees

  • 19th IEEE International Conference on Computational Science and Engineering (CSE 2016), Track Chair: "Parallel Programming for Scientific Computing"

Publications

Matthias Korch; Tim Werner:
An in-depth introduction of multi-workgroup tiling for improving the locality of explicit one-step methods for ODE systems with limited access distance on GPUs
2020, in: Concurrency and Computation: Practice and Experience, Bd.:n/a, Hnr.: n/a, , DOI: 10.1002/cpe.6016 , BIB

Matthias Korch; Tim Werner:
Multi-workgroup Tiling to Improve the Locality of Explicit One-Step Methods for ODE Systems with Limited Access Distance on GPUs
2020, Cham, Springer International Publishing, ISBN/ISSN/ISMV Nummer:978-3-030-43229-4 in: Parallel Processing and Applied Mathematics , BIB

Matthias Korch; Philipp Raithel; Tim Werner:
Implementation and Optimization of a 1D2V PIC Method for Nonlinear Kinetic Models on GPUs
2020, in: , DOI: 10.1109/PDP50117.2020.00012 , BIB

Johannes Seiferth; Matthias Korch; Thomas Rauber:
Offsite Autotuning Approach -- Performance Model Driven Autotuning Applied to Parallel Explicit ODE Methods
2020, Cham, Springer International Publishing, in: High Performance Computing. ISC High Performance 2020 , BIB

Matthias Korch; Tim Werner:
Improving locality of explicit one-step methods on GPUs by tiling across stages and time steps
2020, in: Future Generation Computer Systems, Bd.:102, S.: 889  -  901 , DOI: https://doi.org/10.1016/j.future.2019.07.075 , BIB

Markus Scherg; Johannes Seiferth; Matthias Korch; Thomas Rauber:
Performance Prediction of Explicit ODE Methods on Multi-Core Cluster Systems
2019, New York, NY, USA, Association for Computing Machinery, ISBN/ISSN/ISMV Nummer:9781450362399 in: Proceedings of the 2019 ACM/SPEC International Conference on Performance Engineering , DOI: 10.1145/3297663.3310306 , BIB

Matthias Korch; Tim Werner:
Accelerating explicit ODE methods on GPUs by kernel fusion
2018, in: Concurrency and Computation: Practice and Experience, Bd.:30, Hnr.: 18, , DOI: 10.1002/cpe.4470 , BIB

Matthias Korch; Tim Werner:
Exploiting Limited Access Distance for Kernel Fusion Across the Stages of Explicit One-Step Methods on GPUs
2018, in: Proceedings of the 30th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD 2018) , BIB

Johannes Seiferth; Christie Alappat; Matthias Korch; Thomas Rauber:
Applicability of the ECM Performance Model to Explicit ODE Methods on Current Multi-core Processors
2018, Cham, Springer International Publishing, ISBN/ISSN/ISMV Nummer:978-3-319-92040-5 in: High Performance Computing , BIB

Matthias Korch; Tim Werner:
Efficient GPU-based Smoothed Particle Hydrodynamics
2017, in: International Workshop on Parallel Numerics (PARNUM 2017) , DOI: 10.13140/RG.2.2.21647.48806 , BIB

Matthias Korch; Thomas Rauber; Matthias Stachowski; Tim Werner:
Influence of locality on the scalability of method- and system-parallel explicit peer methods [Awarded]
2016, in: 2016 Federated Conference on Computer Science and Information Systems (FedCSIS) , BIB

Natalia Kalinnik; Matthias Korch; Thomas Rauber:
Online auto-tuning for the time-step-based parallel solution of ODEs on shared-memory systems
2014, in: Journal of Parallel and Distributed Computing , , DOI: 10.1016/j.jpdc.2014.03.006 , BIB

Matthias Korch; Tobias Ramming; Gerhard Rein:
Parallelization of Particle-in-Cell Codes for Nonlinear Kinetic Models from Mathematical Physics
2013, IEEE Computer Society Press, in: Proceedings of the 2013 International Conference on Parallel Processing (ICPP 2013) , note: Short Paper , DOI: 10.1109/ICPP.2013.65 , BIB

Marvin Ferber; Natalia Kalinnik; Matthias Korch; Andreas Prell; Thomas Rauber; Matthias Witzgall:
MAP: Mobile Assistance Platform with a VM Type Selection Ability
2013, in: Proceedings of the 19th IEEE International Conference on Parallel and Distributed Systems (ICPADS) , DOI: 10.1109/ICPADS.2013.79 , BIB

Matthias Korch:
Locality Improvement of Data-Parallel Adams--Bashforth Methods through Block-Based Pipelining of Time Steps
2012, Springer-Verlag, in: Euro-Par 2012. Parallel Processing , DOI: 10.1007/978-3-642-32820-6_56 , BIB

Matthias Korch; Julien Kulbe; Carsten Scholtes:
Diamond-Like Tiling Schemes for Efficient Explicit Euler on GPUs
2012, in: 11th International Symposium on Parallel and Distributed Computing (ISPDC 2012) , note: Best paper award , DOI: 10.1109/ISPDC.2012.42 , BIB

Matthias Korch:
Exploiting Limited Access Distance of ODE Systems for Parallelism and Locality in Explicit Methods
2012, Slovak University of Technology in Bratislava, Faculty of Civil Engineering, Department of Mathematics and Descriptive Geometry, in: ALGORITMY 2012. 19th Conference on Scientific Coputing, Vysoké Tatry -- Podbanské, Slovakia, September 9--14, 2012. Proceedings of contributed papers and posters , BIB

Natalia Kalinnik; Matthias Korch; Thomas Rauber:
An efficient time-step-based self-adaptive algorithm for Predictor-Corrector methods of Runge-Kutta type
2011, in: Journal of Computational and Applied Mathematics. Special Issue CANA 2010, S.: 394  - 410 , DOI: 10.1016/j.cam.2011.07.015 , BIB

Matthias Korch; Thomas Rauber; Carsten Scholtes:
Memory-Intensive Applications on a Many-Core Processor
2011, Los Alamitos, CA, USA, IEEE Computer Society, ISBN/ISSN/ISMV Nummer:978-0-7695-4538-7 in: High Performance Computing and Communications (HPCC), 2011 IEEE 13th International Conference on , DOI: 10.1109/HPCC.2011.26 , BIB

Matthias Korch; Thomas Rauber; Carsten Scholtes:
Scalability and locality of extrapolation methods on large parallel systems
2011, John Wiley & Sons, Ltd., in: Concurrency and Computation: Practice and Experience, Bd.:23, Hnr.: 15, S.: 1789  - 1815 , DOI: 10.1002/cpe.1765 , BIB

Natalia Kalinnik; Matthias Korch; Thomas Rauber:
Dynamic Selection of Implementation Variants of Sequential Iterated Runge-Kutta Methods with Tile Size Sampling
2011, ACM, in: Proceeding of the Second Joint WOSP/SIPEW International Conference on Performance Engineering (ICPE'11) , DOI: 10.1145/1958746.1958775 , BIB

Matthias Korch; Thomas Rauber:
Parallel Low-Storage Runge-Kutta Solvers for ODE Systems with Limited Access Distance
2011, SAGE Publications, in: International Journal of High Performance Computing Applications, Bd.:25, Hnr.: 2, S.: 236  - 255 , DOI: 10.1177/1094342010384418 , BIB

Matthias Korch; Thomas Rauber; Carsten Scholtes:
Mixed-Parallel Implementations of Extrapolation Methods with Reduced Synchronization Overhead for Large Shared-Memory Computers
2010, in: Parallel and Distributed Systems (ICPADS), 2010 IEEE 16th International Conference on , DOI: 10.1109/ICPADS.2010.12 , BIB

Matthias Korch; Thomas Rauber; Carsten Scholtes:
Scalability and Locality of Extrapolation Methods for Distributed-Memory Architectures
2010, Springer-Verlag, in: Euro-Par 2010, Part~II , DOI: 10.1007/978-3-642-15291-7_8 , BIB

Natalia Kalinnik; Matthias Korch; Thomas Rauber:
Applicability of Dynamic Selection of Implementation Variants of Sequential Iterated Runge-Kutta Methods
2010, IEEE Computer Society Press, in: 2010 IEEE International Conference on Cluster Computing -- Workshops and Tutorials , DOI: 10.1109/CLUSTERWKSP.2010.5613097 , BIB

Matthias Korch; Thomas Rauber:
Parallel Low-Storage Runge-Kutta Solvers for ODE Systems with Limited Access Distance
2010, Bayreuth, Germany, , BIB

Matthias Korch; Thomas Rauber:
Storage Space Reduction for the Solution of Systems of Ordinary Differential Equations by Pipelining and Overlapping of Vectors
2010, IEEE Computer Society Press, in: Proceedings of the IPDPS 2010 Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC-10) , DOI: 10.1109/IPDPSW.2010.5470768 , BIB

Matthias Korch; Thomas Rauber:
Parallel Implementation of Runge-Kutta Integrators with Low Storage Requirements
2009, Springer-Verlag, in: Euro-Par 2009. Parallel Processing , DOI: 10.1007/978-3-642-03869-3_73 , BIB

Matthias Korch; Thomas Rauber:
Scalability of Time- and Space-Efficient Embedded Runge-Kutta Solvers for Distributed Address Space
2009, IEEE Computer Society Press, in: Proceedings of the 2009 International Conference on Parallel Processing (ICPP-2009) , DOI: 10.1109/ICPP.2009.14 , BIB

Sascha Hunold; Matthias Korch; B. Krellner; Thomas Rauber; T. Reichel; G. Rünger:
Inkrementelle Transformation einer monolithischen Geschäftssoftware
2008, Bonn, Köllen Druck+Verlag GmbH, in: Software Engineering 2008 -- Workshopband , BIB

Sascha Hunold; Matthias Korch; B. Krellner; Thomas Rauber; T. Reichel; G. Rünger:
Transformation of Legacy Software into Client/Server Applications through Pattern-based Rearchitecturing
2008, IEEE, in: 32nd Annual IEEE International Computer Software and Applications Conference (COMPSAC 2008) , DOI: 10.1109/COMPSAC.2008.158 , BIB

Matthias Korch:
Eingebettete Runge-Kutta-Verfahren für parallele Rechnersysteme: Effiziente Implementierung durch Ausnutzung der Speicherzugriffslokalität
2008, Saarbrücken, Germany, VDM Verlag Dr. Müller, ISBN/ISSN/ISMV Nummer:978-3-8364-9140-2 , BIB

Ralf Hoffmann; Sascha Hunold; Matthias Korch; Thomas Rauber:
Towards Scalable Parallel Numerical Algorithms and Dynamic Load Balancing Strategies
2008, Springer-Verlag, ISBN/ISSN/ISMV Nummer:978-3-540-69181-5 in: High Performance Computing in Science and Engineering, Garching/Munich 2007. Transactions of the Third Joint HLRB and KONWIHR Status and Result Workshop, Dec. 3--4, 2007, Leibniz Supercomputing Centre, Garching/Munich, Germany , DOI: 10.1007/978-3-540-69182-2_40 , BIB

Matthias Korch; Thomas Rauber:
Locality Optimized Shared-Memory Implementations of Iterated Runge-Kutta Methods
2007, Springer-Verlag, in: Euro-Par 2007. Parallel Processing , DOI: 10.1007/978-3-540-74466-5_78 , BIB

Matthias Korch:
Effiziente Implementierung eingebetteter Runge-Kutta-Verfahren durch Ausnutzung der Speicherzugriffslokalität
2006, Smart-Link: http://opus.ub.uni-bayreuth.de/volltexte/2007/280/ , BIB

Matthias Korch; Thomas Rauber:
Optimizing Locality and Scalability of Embedded Runge-Kutta Solvers Using Block-Based Pipelining
2006, Elsevier, in: Journal of Parallel and Distributed Computing, Bd.:66, Hnr.: 3, S.: 444  - 468 , DOI: 10.1016/j.jpdc.2005.09.003 , BIB

Matthias Korch; Thomas Rauber:
Applicability of Load Balancing Strategies to Data-Parallel Embedded Runge-Kutta Integrators
2006, Springer-Verlag, in: Euro-Par 2006. Parallel Processing , DOI: 10.1007/11823285_75 , BIB

Matthias Korch; Thomas Rauber:
Simulation-based analysis of parallel Runge-Kutta solvers
2006, Springer-Verlag, in: Applied Parallel Computing: State of the Art in Scientific Computing. 7th International Workshop, PARA 2004, Lyngby, Denmark, June 2004. Revised Selected Papers , DOI: 10.1007/11558958_133 , BIB

Matthias Korch; Thomas Rauber:
Comparison of Parallel Implementations of Runge-Kutta Solvers: Message Passing vs. Threads
2004, Elsevier, ISBN/ISSN/ISMV Nummer:0-444-51689-1 in: Parallel Computing: Software Technology, Algorithms, Architectures & Applications -- Proceedings of the International Conference ParCo2003, Dresden, Germany , BIB

Ralf Hoffmann; Matthias Korch; Thomas Rauber:
Using Hardware Operations to Reduce the Synchronization Overhead of Task Pools
2004, IEEE, in: Proc. of the Int. Conference on Parallel Processing (ICPP) , DOI: 10.1109/ICPP.2004.1327927 , BIB

Ralf Hoffmann; Matthias Korch; Thomas Rauber:
Performance Evaluation of Task Pools Based on Hardware Synchronization
2004, Pittsburgh, USA, ACM/IEEE, in: Proc. of the Supercomputing 2004 , BIB

Matthias Korch; Thomas Rauber:
A comparison of task pools for dynamic load balancing of irregular algorithms
2004, John Wiley & Sons, in: Concurrency and Computation: Practice and Experience, Bd.:16, Hnr.: 1, S.: 1  - 47 , DOI: 10.1002/cpe.745 , BIB

Matthias Korch; Thomas Rauber; Gudula Rünger:
Performance Optimization of RK Methods Using Block-based Pipelining
2003, Kluwer Academic Publishers, in: Performance Analysis and Grid Computing , BIB

Matthias Korch; Thomas Rauber:
Scalable Parallel RK Solvers for ODEs Derived by the Method of Lines
2003, Springer-Verlag, in: Euro-Par 2003. Parallel Processing , BIB

Matthias Korch; Thomas Rauber:
Evaluation of Task Pools for the Implementation of Parallel Irregular Algorithms
2002, IEEE Computer Society Press, in: Proceedings of the 2002 ICPP Workshops , DOI: 10.1109/ICPPW.2002.1039782 , BIB

Matthias Korch; Thomas Rauber; Gudula Rünger:
Pipelining for Locality Improvement in RK Methods
2002, Springer-Verlag, in: Euro-Par 2002. Parallel Processing , BIB

Matthias Korch:
Einsatz von Taskpools in Pthreads und Java zur parallelen Implementierung irregulärer Algorithmen
2001, , BIB

University of Bayreuth -