Sebastiano Fabio Schifano has spent the early scientific career at CNR (Italian National Research Council) first, and at INFN (Italian National Institute for Nuclear Physics) after, as researcher and senior researcher. After, he moved to University of Ferrara as assistant professor of operating systems. The research activities are mainly focused on the design and exploitation of high performance computing systems for scientific applications, like Lattice Gauge Theory (LQCD), fluid-dynamics and spin-glasses.
During the period at INFN he had a major role in the development of the APEmille and apeNEXT massively parallel systems, and he has been deeply involved in the design of the processor architecture and the network communication system. Within the framework of the APE projects he has been responsible also for the development of the back-end steps of the compilation chain to optimize the scheduling of VLIW micro-instructions, and of the operating system.
In 2005 he was co-autor of the proposal of the Janus project, an heterogeneous massively parallel system based on FPGAs, developed mainly as spin-glass simulation engine, but it can be easily reconfigured for others applications. In the framework of this project he was responsible for the design of the architecture of the system, and he coordinated the development of the interface between the boards and the host. In 2014 a second generation of this machine has been developed using more recent FPGA devices.
In 2007 he was co-autor of the proposal for the QPACE project, a German project to develop a massively parallel system based on IBM Cell-BE processors, interconnected by a custom 3D-mesh. In the framework of the QPACE project he was in charge of designing and developing the logic of the network processor implemented on FPGA. In November 2009 and June 2010 the QPACE machine has been awarded as the top entry of the GREEN500 list (www.green500.org).More recently he has been involved in several EU projects, like Hadron Physics 2 and Hadron Physics 3, TORUS, STIMULATE and EUROEXA.
Main research interests are in the field of parallel programming, code optimization, design of parallel systems and performance analysis of processors.