The Santos Dumont Supercomputing System and the National High-Performance Computing System's (SINAPAD) role in HPC research and general Brazilian science development

High-performance computing (HPC) platforms are required to solve the most diverse large-scale scientific problems in various research areas, such as biology, chemistry, physics, and health sciences. Researchers use a multitude of scientific software, which have different requirements. Thus, supercomputers must efficiently handle mixed workloads when storing application data. The Santos Dumont Supercomputer (SDumont), a Bull/Atos machine located at the National Laboratory for Scientific Computing (LNCC) in Brazil, is an example of an HPC environment with significant heterogeneity of research on several areas of knowledge, each with a specific set of scientific software. The primary ones are Chemistry (21.3%), Physics (17.1%), Engineering (12.6%), and Biological Sciences (10,1%). SDumont is the Tier-0 of the National High-Performance Computing System (SINAPAD) and one of the largest in Latin America. It currently has 233 research projects in progress, with a daily average of ≈ 300 concurrent jobs. In operation since 2016, SDumont has had 18,424 CPU cores distributed across 758 compute nodes (CN), corresponding to 5 PFlops total computing power. An upgrade to 25 Ptops is expected to be installed by October 2024. This talk will discuss the Santos Dumont Supercomputing System and the National High-Performance Computing System's (SINAPAD) role in HPC research and general Brazilian science development. www.lncc.br https://sdumont.lncc.br/ https://www.lncc.br/sinapad/