Jag är doktorand vid avdelningen för teknisk databehandling (TDB), med Sverker Holmgren som huvudhandledare.
Forskning
Min forskning riktar sig mot utveckling och implementation av storskaliga adaptiva algoritmer, framförallt för högdimensionella tidsberoende partiella differentialekvationer. Jag är även intresserad av kommunikationsminimerande algoritmer för numerisk linjär algebra.
- HAParaNDA, ett adaptivt parallellt ramverk för högdimensionella tidsberoende PDE-problem
- Numerisk kvantdynamik
- Projekt inom UPMARC
Publikationer
Granskade publikationer
- Data structures and algorithms for high-dimensional structured adaptive mesh refinement. I Advances in Engineering Software, volym 82, pp 75-86, 2015. (DOI).
- Stable difference methods for block-oriented adaptive grids. I Journal of Scientific Computing, volym 65, pp 486-511, 2015. (DOI).
- Communication-efficient algorithms for numerical quantum dynamics. I Applied Parallel and Scientific Computing: Part II, volym 7134 av Lecture Notes in Computer Science, pp 368-378, Springer-Verlag, Berlin, 2012. (DOI).
- An implementation framework for solving high-dimensional PDEs on massively parallel computers. I Numerical Mathematics and Advanced Applications: 2009, pp 417-424, Springer-Verlag, Berlin, 2010. (DOI).
Andra artiklar och rapporter
- Adaptive Solvers for High-Dimensional PDE Problems on Clusters of Multicore Processors. Doktorsavhandling, Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology nr 1199, Acta Universitatis Upsaliensis, Uppsala, 2014. (fulltext, preview image).
- Parallel data structures and algorithms for high-dimensional structured adaptive mesh refinement. Technical report / Department of Information Technology, Uppsala University nr 2014-020, 2014. (Extern länk).
- Data structures and algorithms for high-dimensional structured adaptive mesh refinement. Technical report / Department of Information Technology, Uppsala University nr 2014-019, 2014. (Extern länk).
- Towards an adaptive solver for high-dimensional PDE problems on clusters of multicore processors. Licentiatavhandling, IT licentiate theses / Uppsala University, Department of Information Technology nr 2012-003, Uppsala University, 2012. (fulltext).
- Numerical evaluation of the Communication-Avoiding Lanczos algorithm. Technical report / Department of Information Technology, Uppsala University nr 2012-001, 2012. (Extern länk).
- Stable difference methods for block-structured adaptive grids. Technical report / Department of Information Technology, Uppsala University nr 2011-022, 2011. (Extern länk).
- Efficient implementation of a high-dimensional PDE-solver on multicore processors. I Proc. 2nd Swedish Workshop on Multi-Core Computing, pp 64-66, Department of Information Technology, Uppsala University, Uppsala, Sweden, 2009.
Presentationer
- M. Gustafsson: Towards an adaptive solver for high-dimensional PDE problems on clusters of multicore processors, Licentiate seminar, Uppsala, March 2012.
- M. Gustafsson, K. Kormann, S. Holmgren: Communication-efficient Krylov methods for exponential integration in quantum dynamics, PARA 2010, Reykjavík, June 2010.
- M. Gustafsson, K. Kormann, S. Holmgren: Parallel Lanczos-based exponential integrators for quantum dynamics, CMCIM 2010, Copper Mountain, April 2010.
- M. Gustafsson, S. Holmgren: Multicore-tailored implementation of s-step methods for high-dimensional PDEs, SIAM PP 2010, Seattle, February 2010.
- M. Gustafsson, S. Holmgren: Optimizing high-dimensional stencil operators for multicore processors, MCC 2009, Uppsala, November 2009.
- M. Gustafsson, S. Holmgren: An implementation framework for solving high-dimensional PDEs on massively parallel computers, ENUMATH 2009, Uppsala, June 2009.
Postrar
- M. Gustafsson, A. Nissen, K. Kormann: High-Dimensional Adaptive Grids for Time-Dependent PDEs, SIAM CSE, Reno, March 2011.
Undervisning
Kurser jag har undervisat:
- Programmering av Parallelldatorer (VT 2009, VT 2011, VT 2012)
- Beräkningsvetenskap I (VT 2009, VT 2010)