About Me

I am a Research Scientist working in the Performance group at Innovative Computing Laboratory (ICL), University of Tennessee Knoxville. Before joining ICL in July 2021 I was a Postdoctoral Fellow at Argonne National Laboratory (ANL), where I worked in the Programming Models and Runtime Systems group. Before ANL I also worked in industry, at Xyratex and Seagate, while persuing my Ph.D. in computer science at the University of Mainz in Germany.

During my years at Xyratex and Seagate I was involved in several EU funded Exascale Computing projects such as DEEP-ER, SAGE and EsiWace. A common theme for all these projects was the improvement of I/O performance for scientific workloads. This was achieved, in part, by re-engineering existing I/O middlewares and libraries. In DEEP-ER this resulted into the extension of the MPI-IO hints mechanism to take advantage of non-volatile memory devices locally attached to compute nodes in a HPC cluster, effectively adding a new storage layer between main memory and the parallel file system that could be used as burst buffer for checkpoint operations.

At Argonne National Laboratory I was involved in the development of new features for the MPICH Project, implementation of the MPI standard. These included the integration and support of new memory technologies (e.g. HBM) and GPU accelerators.

Currently, I am involved in the PAPI Project. PAPI is a software library allowing users to tap into the hardware performance counters, nowadays present in all major hardware components (ranging from CPU to network cards), to understand and improve the interaction between hardware and software. PAPI is the industry standard for hardware performance monitoring and is used by many other successful packages such as HPCToolkit and Tau.