This research group has several lines of work. The most important are the following:
The development of efficient algorithms and software to tackle NLA operations, from the perspective of high performance computing (HPC). Specifically, leveraging massively parallel platforms, such as graphics processing units (GPUs).
The application of different HPC techniques to enhance the computational performance of large numerical models that simulate, for example, the state of the atmosphere or the behavior of waterbodies.
Some of our most representative publications
Ernesto Dufrechou, Pablo Ezzatti: A New GPU Algorithm to Compute a Level Set-Based Analysis for the Parallel Solution of Sparse Triangular Systems. IPDPS 2018: 920-929
Rodrigo Baya, Claudio Porrini, Martín Pedemonte, Pablo Ezzatti: Task Parallelism in the WRF Model Through Computation Offloading to Many-Core Devices. PDP 2018: 596-600
Martín Pedemonte, Francisco Luna, Enrique Alba: A theoretical and empirical study of the trajectories of solutions on the grid of Systolic Genetic Search. Inf. Sci. 445-446: 97-117 (2018)
Peter Benner, Pablo Ezzatti, Enrique S. Quintana-Ortí, Alfredo Remón: Extending the Gauss-Huard method for the solution of Lyapunov matrix equations and matrix inversion. Concurrency and Computation: Practice and Experience 29(9) (2017)
Ignacio Decia, Rodrigo Leira, Martín Pedemonte, Eduardo Fernández, Pablo Ezzatti: A VNS with Parallel Evaluation of Solutions for the Inverse Lighting Problem. EvoApplications (1) 2017: 741-756
Juan Pablo Silva, Ernesto Dufrechou, Pablo Ezzatti, Enrique S. Quintana-Ortí, Alfredo Remón, Peter Benner: Balancing Energy and Performance in Dense Linear System Solvers for Hybrid ARM+GPU platforms. CLEI Electron. J. 19(1): 2 (2016)
We collaborate with other groups worldwide. Some of them are the following:
Julio Herrera y Reissig 565, Montevideo, Uruguay
+598 2711 42 44
mpedemon [at] fing [dot] edu [dot] uy