Allan P. Engsig-Karup
Associate Professor in Scientific Computing, M.Sc.(Eng.), Ph.D.
Associate Professor Allan P. Engsig-Karup recieved an M.Sc. in Coastal engineering and applied math from the Technical University of Denmark (DTU) in January 2003. During the studies, 6 months of 2001 was spend as an international student at Adelaide University, Adelaide, Australia. Following graduation he was awarded a three year fellowship to begin work towards a Ph.D. at DTU at Coastal, Maritime and Structural Engineering Section in the Department of Mechanical Engineering, DTU.
During the three years of PhD study he worked on the development of a novel advanced solution strategy of the so-called high-order Boussinessq-type equations using the Discontinous Galerking Finite Elemenent Methods (e.g. see DGFEM2009), a seven month visit abroad was spend in the period 2004-2005 in the Division of Applied Mathematics at Brown University, Providence, USA, and a one month visit a Rice University, Texas, USA. In August 2006, he recieved a Ph.D. in Coastal engineering and applied math from the Department of Mechanical Engineering, DTU, Denmark.
Following graduation in August 2006, he was awarded a Postdoctoral Fellowship from the Danish Technical Research Council for research in coastal engineering and applied math. A new efficient and robust multigrid preconditioned solution strategy was developed and proposed for simulation of unsteady nonlinear free surface wave over uneven bottoms in three space dimensions (OceanWave3D). The mathematical basis is a unified (in terms of dispersion and nonlinear properties) potential flow model. The numerical solver was developed using a classical flexible-order finite diference method and was tailored to be both robust, efficient, accurate and scalable with respect to computational effort. Currently, I am collaborating with the Computational Hydraulics Group at DTU Mechanics on improving and extending these models further. The most recent achievement is a low-storage accelerated version of this model using Graphics Processing Units (GPUs) for massively parallel computations.
As of mid August 2008, he was appointed Assistant Professor of Scientific Computing at Section for Scientific Computing, Department of Informatics and Mathematical Modeling, DTU, Denmark. In this position the teaching and research focus will be on numerical techniques for solving ordinary and partial differential equations with special interests related to compute-intensive time-dependent problems that may benefit from using high-performance computing ressources (see some related activities in the competence center GPULAB). Developing fast engineering tools for high-performance computing on modern many-core architectures is of key interest and for large problems requires efficient algoritms for efficient and scalable solution of linear systems. This requires tailoring algorithms and solvers to problems and specific needs for best performance (e.g. see ITSOL2011).
In July 2011, he was promoted to Associate Profesor at Section for Scientific Computing, Department of Informatics and Mathematical Modeling, DTU, Denmark.