Alberto F. martÃn
Assistant Research Professor
Computer Science
Pere Virgili Health Research Institute
Spain
Biography
Dr. Alberto F. Martín is a post-doc researcher at CIMNE & UPC. His background includes a 5-year bachelor degree in Computer Science, and a doctoral degree in Computer Science, the latter with specialization on High Performance Scientific Computing. With more than six years of experience in the design and development of scalable parallel preconditioning codes for sparse linear systems arising from the Finite Element discretization of Partial Differential Equations, he is currently co-author and an active developer of FEMPAR since he joined the LSSC team in 2011. FEMPAR is a massively parallel finite element software distributed as free software under the terms of the GNU-GPL v3 license. FEMPAR has been positioned as world's most scalable domain decomposition solver, and it has been proven to efficiently exploit the whole JUQUEEN supercomputer (the largest supercomputer in Europe), a 28-rack IBM BG/Q system equipped with 458,752 cores. FEMPAR is now in the High-Q club of the most scalable codes on JUQUEEN. His main research activities are particularly focused on the following topics: Parallel scientific computing; Multi-threaded and message-passing algorithms for sparse numerical linear algebra; Parallel multilevel preconditioning and iterative solution of sparse linear systems; Large-scale implementation of domain decomposition methods; Parallel programming models, libraries and software for current multicore-based massively parallel computers.
Research Interest
His main research activities are particularly focused on the following topics: Parallel scientific computing; Multi-threaded and message-passing algorithms for sparse numerical linear algebra; Parallel multilevel preconditioning and iterative solution of sparse linear systems; Large-scale implementation of domain decomposition methods; Parallel programming models, libraries and software for current multicore-based massively parallel computers.
Publications
-
Badia S, MartÃn AF, Principe J. Multilevel balancing domain decomposition at extreme scales. SIAM Journal on Scientific Computing. 2016 Jan 26;38(1):C22-52.
-
Badia S, MartÃn AF, Principe J. A highly scalable parallel implementation of balancing domain decomposition by constraints. SIAM Journal on Scientific Computing. 2014 Apr 10;36(2):C190-218.
-
Aliaga JI, Bollhöfer M, Martı AF, Quintana-Ortı ES. Exploiting thread-level parallelism in the iterative solution of sparse linear systems. Parallel Computing. 2011 Mar 31;37(3):183-202.
