Author: Luca Stornaiuolo

Author(s) Luca Stornaiuolo, Marco Santambrogio, Donatella Sciuto Conference2018 IEEE Computer Society Annual Symposium on VLSI (ISVLSI) AbstractDeep Learning algorithms are gaining momentum as main components in a large number of fields, from computer vision and robotics to finance and biotechnology. At the same time, the use of Field Programmable Gate Arrays (FPGAs) for data-intensive applications is increasingly widespread thanks to the possibility to customize hardware accelerators and achieve high-performance implementations with low energy consumption. Moreover, FPGAs have demonstrated to be a viable alternative to GPUs in embedded systems applications, where the benefits of the reconfigurability properties make the system more robust, capable to face the system failures and to respect the constraints of the embedded devices. In this work, we present a framework to efficiently implement Deep Learning algorithms by…

Author(s) Luca Stornaiuolo, Alberto Parravicini, Donatella Sciuto, Marco D Santambrogio Conference 2018 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW) Abstract Hardware accelerators are an effective solution to increase the performance of algorithms in a wide array of disciplines, from data science to computational finance. However, data scientists and mathematicians often do not have the required knowledge or time to fully exploit these accelerators, and they perceive them as difficult and frustrating to use. OpenCL was created to simplify the creation of computational pipelines with heterogeneous hardware, but as of today, its integration with high-level languages commonly used in data science is limited. In this paper, we propose a framework to integrate OpenCL kernels running on Field Programmable Gate Arrays (FPGAs) with Python, R, and MATLAB, the most common…

Author(s) Lorenzo Di Tucci, Marco Rabozzi, Luca Stornaiuolo, Marco D Santambrogio Conference 2017 IEEE International Conference on Computer Design (ICCD) Abstract In the context of heterogneous computing, even though GPUs are the components of election due to both their intrinsically parallel nature and their flexibility, FPGAs are being investigated and experimented due to superior power efficiency on selected workloads While GPUs are the heterogeneous components of election due to both their intrinsically parallel nature and their flexibility, FPGAs are being investigated and experimented due to superior power efficiency on selected workloads. However, the lack of adequate languages, runtimes, programming flexibility and, broadly speaking, proven system level approaches for FPGA-accelerated applications are the most relevant limiting factors to the adoption of these devices into mainstream. In these regards, Amazon recently released…