Experienced researcher in several international projects focused on computational science and engineering. Skilled in Parallel Programming(Multithreaded, Distributed, Hybrid; OpenMP, MPI, CUDA), Numerical Analysis and Linear Algebra, Data Analysis, and Linux.
Managing in-house GPU servers (tiny cluster) and work-stations. Responsible for organizing and conducting seminars/trainings. Researcher in international projects, and supporting teams regarding computation.
Worked with academic members of the university in the projects from academic to industrial. Responsible to assist lectures and training courses about scientific computation and parallel programming (MPI, OpenMP and CUDA).
Responsible for maintaining IT serversand HPC (High Performance Computing) resources.
Thesis Advisor: Prof. Dr. Ahmet Duran
Thesis Advisor: Prof. Dr. M. Serdar Çelebi
Researcher, Parallel Algorithm (Kernel) Development for Large Scale Sparse Linear Systems in Oil Reservoir Simulation (PASSOR), Computational Linear Algebra Project for Saudi ARAMCO, Jul. 2012 - Jun. 2017
Researcher, HPC-based Design of a Novel Electromagnetic Stirrer for Steel Segment Casting, in EU Horizon 2020 – PRACE 4IP (The Partnership for Advanced Computing in Europe – 4th Implementation Phase Project, persistent pan-European Research Infrastructure for High Performance Computing), Feb. 2015 - Apr. 2016
Researcher, Scalable Parallel Nonlinear Parameter Optimization Algorithm with Parameter Pools, PRACE-2IP WP12 T12.2, Sept. 2013 - Aug. 2014
Researcher, Scalability of OpenFOAM for Bio-medical Flow Simulations, PRACE-3IP WP7 T7.2, Sept. 2012 - Aug. 2014
Researcher, Analysis of SuperLU Solvers on the Intel MIC Architecture, PRACE-1IP WP7 T7.1, Mar. 2013 - Dec. 2013
Researcher, Scalable and improved SuperLU on GPU for heterogeneous systems, PRACE-2IP WP12 T12.2, Sept. 2011 - Aug. 2013
Researcher, Performance Analysis of BLAS Libraries in SuperLU_DIST for SuperLU_MCDT (Multi Core Distributed) Development, PRACE-2IP WP12 T12.2, Sept. 2011 - Aug. 2013
Researcher, Structural Analysis of Large Sparse Matrices for Scalable Direct Solvers, PRACE-2IP WP12 T12.2, Sept. 2011 - Aug. 2013
Researcher, Design and Implementation of New Hybrid Algorithm and Solver on CPU for Large Sparse Linear Systems, PRACE-2IP WP12 T12.2, Sept. 2011 - Aug. 2013
M. Tuncel and A. Duran, Effectiveness of grid and random approaches for a model parameter vector optimization, Journal of Computational Science, Vol. 67, 2023, 101960, Doi: 10.1016/j.jocs.2023.101960
M. S. Celebi, A. Duran, F. O. Topkaya, M. Tuncel, B. Akaydin (2017). On the improvement of a scalable sparse direct solver for unsymmetrical linear equations. The Journal of Supercomputing, 73(5), 1852-1904., Doi: 10.1007/s11227-016-1892-7
A. Duran, M.S. Celebi, S. Piskin, and M. Tuncel, (2015). Scalability of OpenFOAM for bio-medical flow simulations. The Journal of Supercomputing, 71(3), 938-951, Doi: 10.1007/s11227-014-1344-1
A. Duran and M. Tuncel, Evaluation of a new parallel numerical parameter optimization algorithm for a dynamical system, Proceedings of the 2nd International Conference Numerical Computations: Theory and Algorithms (NUMTA2016), Italy, 19-25 June 2016, AIP Conference Proceedings 1776, 090052 (2016), Doi: 10.1063/1.4965416
A. Duran, S. Piskin, and M. Tuncel, Evaluating the maturity of OpenFOAM simulations on GPGPU for bio-fluid applications, Proceedings of the Emerging Technology (EMiT) Conference, pp. 11-14, Barcelona Supercomputing Center, Spain, 2-3 June 2016, editors: B.D.Rogers, D.Topping, F.Mantovani, M.K.Bane. ISBN 978-0-9933426-3-9
M. Tuncel, A. Duran, M. S. Celebi, B. Akaydin, and F. O. Topkaya, A comparison of SuperLU solvers on the intel mic architecture, Proceedings of the 2nd International Conference Numerical Computations: Theory and Algorithms (NUMTA2016), Italy, 19-25 June 2016, AIP Conference Proceedings 1776, 090030 (2016), Doi: 10.1063/1.4965394
I. Mazza, A. Duran, Y. Hundur, C. Persi, A. Santoro, and M. Tuncel, Scalability of OpenFOAM for Simulations of a Novel Electromagnetic Stirrer for Steel Casting, Proceedings of the 2016 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA'16), World Congress in Computer Science, Computer Engineering & Applied Computing, pp. 111-116, ISBN: 1-60132-444-8, CSREA Press, July 25-28, 2016, Las Vegas, USA
A. Duran and M. Tuncel, Spectral Effects Of Large Matrices From Oil Reservoir Simulators On Performance Of Scalable Direct Solvers, OnePetro, 172984 SPE Paper – 2014
A. Duran, M.S. Celebi, M. Tuncel and F. Oztoprak, Spectral Analysis of Large Sparse Matrices for Scalable Direct Solvers, Advances in Applied Mathematics, Springer Proceedings in Mathematics & Statistics 87, pp. 153-160, 2014. Doi: 10.1007/978-3-319-06923-4
A. Duran, M.S. Celebi, M. Tuncel, Scalability of SuperLU Solvers for Large Scale Complex Reservoir Simulations, Int. Conference for Mathematical Methods in Fluid Dynamics and Simulation of Giant Oil and Gas Reservoirs, Joint SPE and SIAM Conf., Istanbul-Turkey, Sept. 3-5, 2012
Technical Support, 6th National HPC and Parallel Programming Summer School, Istanbul Technical University, 4 - 15 June 2012
Attended, PRACE Autumn School co-organized by CEA and GENCI, Bruyères-le-Châtel, France, 25 - 27 October 2011
PRACE Autumn School focused on advanced hybrid programming and benefit in exclusivity from this brand new CURIE hybrid GPU partition for hands on.
www.prace-ri.eu/PRACE-Autumn-School-2011-25-27
Attended, GPGPU Programming Workshop, Istanbul Technical University, Informatics Institute, 27 - 30 June 2011