Our research group performs research of computing problems whose computational or memory complexity requires deployment of many processors or stand-alone computers. We focus on the modern parallel and distributed computing systems consisting of thousands of computing nodes. We are interested both in data structures and algorithms in selected application domains and in research of system architectures of large-scale high-performance computing systems, such as computational grids, computer clusters, and massively parallel machines.


  • Algorithms for GPGPU computing platforms
  • Architecture of non-dedicated computing clusters
  • Distributed algorithms for P2P clusters
  • Distributed task scheduling in computational grids
  • High-performance computing with very large sparse matrices
  • Memory and storage representation of sparse matrices
  • Multithreaded microarchitectures




P202/12/2011 (Grant Agency of the Czech Republic): Parallel input/output algorithms for large-scale sparse matrices, main investigator: prof. Ing. Pavel Tvrdík, CSc., 01/2012-12/2014.
TA02011394 (Technology Agency of the Czech Republic): Datacenter Management System, main investigator: co-investigator: prof. Ing. Pavel Tvrdík, CSc., 01/2012-12/2013.

LANGR, D., ŠIMEČEK, I., TVRDÍK, P., and DYTRYCH, T.: Large-Scale Visualization of Sparse Matrices. Scalable Computing: Practice and Experience, 2014, 15(1), 21-31. ISSN 1895-1767.
LANGR, D., ŠIMEČEK, I., TVRDÍK, P., and DYTRYCH, T.:  Scalable Parallel Generation of Very Large Sparse Benchmark Matrices. In: Parallel Processing and Applied Mathematics,  Springer, 2014, LNCS 8384, pp. 178-187. ISBN 978-3-642-55223-6.
LANGR, D., TVRDÍK, P., ŠIMEČEK, I., and DYTRYCH, T.: Downsampling Algorithms for Large Sparse Matrices. International Journal of Parallel Programming, 2014, ISSN 0885-7458. Published on-line.
LANGR, D., TVRDÍK, P., DYTRYCH, T., and DRAAYER, J.P.: Paraperm: Parallel generation of random permutations with MPI. ACM Transactions on Mathematical Software, 2014. Published on-line.
ŠIMEČEK, I., LANGR, D., and TVRDÍK, P.: Tree-based Space Efficient Formats for Storing the Structure of Sparse Matrices. Scalable Computing: Practice and Experience, 2014, 15(1), 1-20. ISSN 1895-1767.
GATTERMAYER, J. and TVRDÍK, P.: Different Approaches to Distributed Compilation. In: Parallel and Distributed Processing Symposium Workshops & PhD Forum (IPDPSW), IEEE 26th International, IEEE, 2012, pp. 1128-1134. ISBN 978-1-4673-0974-5.
GATTERMAYER, J. and TVRDÍK, J.: Porting Clondike to Heterogeneous Platforms. In: 2012 2nd IEEE International Conference On Parallel, Distributed and Grid Computing, IEEE, 2012, pp. 380-384. ISBN 978-1-4673-2922-4.
ŠIMEČEK, I., LANGR, D., and TVRDÍK, P.: Minimal Quadtree Format for Compression of Sparse Matrices Storage. In: Proceedings of 14th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC 2012), IEEE, 2012, pp. 359-364. ISBN 978-0-7695-4934-7.
BUŠ, L. and TVRDÍK, P.: Towards auction algorithms for large dense assignment problems. Computational Optimization and Applications, 2009, 43(3), 411-436. ISSN 0926-6003.
ŠŤAVA, M. and TVRDÍK, P.: Security System for Overlapping Non-dedicated Clusters. In: 2009 IEEE International Symposium on Parallel and Distributed Processing with Applications, IEEE, 2009, pp. 272-281. ISBN 978-0-7695-3747-4.

Prof. Ing. Pavel Tvrdík, CSc.

Last modified: 21.10.2014, 14:13