Via Marco Biagi 5 – 73100 Lecce, Italy
(+39) 0832 1902411
Silvia Mocavero works as scientific collaborator at the “Advanced Scientific Computing” (ASC) Division of the “Euro-Mediterranean Centre on Climate Change” (CMCC), where she led the “High End Computing” research group until mid 2018.
She holds a Ph.D. in Innovative Materials and Technologies, obtained in 2006 from ISUFI at the University of Lecce (Italy). In 2010, she was visiting researcher at the Barcelona Supercomputing Centre (BSC) within the HPC-Europa2 initiative for the optimization of the Biogeochemical Flux Model, coupled with NEMO. In 2013, she was visiting scientist at the Argonne National Laboratory (ANL) of Chicago for the porting, scalability analysis and improvement of the NEMO oceanic model on BG/Q architecture.
Her expertise concerns high performance, grid & cloud computing. Her skills include parallel programming on HPC systems and distributed environments, with a deep experience of several parallel programming models such as message passing, shared memory, many-threads programming. Since 2011, she has been exploring new issues related to exascale computing. She works on the analysis and optimization of climate models with a particular focus on NEMO as member of the System Team and the HPC group of the NEMO Consortium. She is currently involved in several H2020 projects.
ULTIME PUBBLICAZIONI
- Parallel implementation of the SHYFEM (System of HydrodYnamic Finite Element Modules) model
- A Bioinfomatics Grid Alignment Toolkit
- Prototype of grid environment for earth system models
- The Roofline Model for Oceanic Climate Applications
- The Performance Model of an Enhanced Parallel Algorithm for the SOR Method
- The performance model for a parallel SOR algorithm using the red-black scheme
- CPMIP: measurements of real computational performance of Earth system models in CMIP6
- Hybridisation strategies and data structures for the NEMO ocean model
- Performance and results of the high-resolution biogeochemical model PELAGOS025 v1.0 within NEMO v3.4
- Performance analysis of the COSMO-CLM model