Computing High Performance Computing |
Sergio Rivas-Gomez is an HPC Performance Engineer in the Computing Division.
Working in the HPC team, Sergio analyzes and optimizes the source code of the different software components utilized in the Project. In particular, he aims at understanding how the I/O performance can be improved by re-designing the access pattern utilized during complex simulations, as well as researching how novel storage technologies (e.g., Infinite Memory Engine) could be seamlessly integrated, among other responsibilities. Before joining the Blue Brain Project, Sergio worked for several years at Microsoft after winning Microsoft’s Imagine Cup Spain in 2013, where he presented a project that used machine learning to analyze the electroencephalography (EEG) of the user. More recently, he was involved in research and pursued a PhD in High-Performance Computing at KTH Royal Institute of Technology in Sweden. During this period, he contributed to two of the leading Exascale Horizon 2020 projects of the European Union (SAGE/Sage2 and EPiGRAM-HS), and collaborated with institutions such as the Pacific Northwest National Laboratory or the National Center for Atmospheric Research in the United States. His research to date, centers on programming models and interfaces for I/O that seamlessly expose novel storage technologies (e.g., NVDIMM), in preparation for the next-generation supercomputers. In addition to his PhD, Sergio has an MSc in High-Performance Computing and a BSc in Computer Science Engineering, from the University of Murcia, Spain. In 2019, he won the Best Paper Award at IEEE HiPC’19 and the Best Paper Award at EuroMPI/USA in 2017. In his free time, he enjoys time with his partner, family, and their beautiful animals.
Publications Sergio Rivas-Gomez, Alessandro Fanfarillo, Sebastién Valat, Christophe Laferriere, Philippe Couvee, Sai Narasimhamurthy, and Stefano Markidis. uMMAP-IO: User-level Memory-mapped I/O for HPC Proceedings of the 26th IEEE International Conference on High- Performance Computing, Data, and Analytics (HiPC’19), IEEE, 2019.
Sergio Rivas-Gomez, Sai Narasimhamurthy, Keeran Brabazon, Oliver Perks, Erwin Laure, and Stefano Markidis. Decoupled Strategy for Imbalanced Workloads in Map-Reduce Frameworks Proceedings of the 20th International Conference on High Performance Computing and Communications (HPCC’18), pp. 921–927. IEEE, 2018. |