Osni Marques será um dos palestrantes no PUBTechSF@Berkeley 2016. O título da sua palestra é “Álgebra linear numérica e seu uso no dia-a-dia“.
O PUBTechSF@Berkeley 2016 será no dia 22 de setembro de 2016, no Berkeley Institute for Data Science (BIDS).
Para reservar seu lugar, inscreva-se aqui!
Essa palestra mostrará um conjunto de problemas que ilustram a utilização da álgebra linear numérica na solução de problemas cotidianos – dois quais a audiência certamente se beneficia, possivelmente à revelia. A palestra procurará também mostrar os efeitos de computadores atuais e de um futuro próximo na solução de tais problemas.
Osni Marques is a member of the Scalable Solvers Group, Computational Research Division, at the Lawrence Berkeley National Laboratory (LBNL), which he joined in 1996. He has worked with high-performance, robust software libraries and tools, that can simplify the development of engineering and scientific application codes, and that provide solutions to common and important computational problems. Osni has studied and implemented algorithms for the solution of problems in numerical linear algebra, in particular eigenvalue problems, that have been used in applications related to protein motions, acoustics problems in automobile design, structural analyses, and inverse problems in geophysics applications. He has collaborated in a project funded by the Department of Energy to study electronic properties of 3D, million-atom semiconductor nanostructures. In this project, he focused on algorithms for the computation of interior eigenstates of quantum dots and nanowires. For many years, Osni has been involved in projects funded by the NSF for further development of the well-known LAPACK and ScaLAPACK libraries of algorithms for dense linear algebra calculations.