Giovanni Brignone

Giovanni Brignone

Auto-generated profile, based on publicly available information

Giovanni Brignone

Department of Electronics and Telecommunications, Politecnico di Torino, Turin, Italy

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 November 2023 in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
Verified
Journal Article
Closed access
FOFabrizio Ottati
Fabrizio Ottati
CGChang Gao
Chang Gao
QCQinyu Chen
Qinyu Chen
GBGiovanni Brignone
Giovanni Brignone
Mario R. CasuMario R. Casu
Mario R. Casu
Luciano LavagnoLuciano Lavagno
Luciano Lavagno
As deep learning models scale, they become increasingly competitive from domains spanning from computer vision to natural language processing; however, this happens at the expense of efficiency since they require increasingly more memory and computing power. The power efficiency of the biological brain outperforms any large-scale deep learning (DL) model; thus, neuromorphic computing tries to mimic the brain operations, such as spike-based information processing, to improve the efficiency of DL models. Despite the benefits of the brain, such as efficient information transmission, dense neuronal interconnects, and the co-location of computation and memory, the available biological substrate has severely constrained the evolution of biological brains. Electronic hardware does not have the same constraints; therefore, while modeling spiking neural networks (SNNs) might uncover one piece of the puzzle, the design of efficient hardware backends for SNNs needs further investigation, potentially taking inspiration from the available work done on the artificial neural networks (ANNs) side. As such, when is it wise to look at the brain while designing new hardware, and when should it be ignored? To answer this question, we quantitatively compare the digital hardware acceleration techniques and platforms of ANNs and SNNs. As a result, we provide the following insights: (i) ANNs currently process static data more efficiently, (ii) applications targeting data produced by neuromorphic sensors, such as event-based cameras and silicon cochleas, need more investigation since the behavior of these sensors might naturally fit the SNN paradigm, and (iii) hybrid approaches combining SNNs and ANNs might lead to the best solutions and should be investigated further at the hardware level, accounting for both efficiency and loss optimization.

4 November 2022 in IEEE Access
Verified
Journal Article
Open access
GBGiovanni Brignone
Giovanni Brignone
MJM. Usman Jamal
M. Usman Jamal
Mihai T. LazarescuMihai T. Lazarescu
Mihai T. Lazarescu
Luciano LavagnoLuciano Lavagno
Luciano Lavagno
Designs implemented on field-programmable gate arrays (FPGAs) via high-level synthesis (HLS) suffer from off-chip memory latency and bandwidth bottlenecks. FPGAs can access both large but slow off-chip memories (DRAM), and fast but small on-chip memories (block RAMs and registers). HLS tools allow exploiting the memory hierarchy in a scratchpad-like fashion, requring a significant manual effort. We propose an automation of the FPGA memory management in Xilinx Vitis HLS through a fully-configurable C++ source-level cache. Each DRAM-mapped array can be associated with a private level 2 (L2) cache with one or more ports, and each port can optionally provide a level 1 cache. The L2 cache runs in a separate dataflow task with respect to the application accessing it. This solution isolates off-chip memory accesses and data buffering into dedicated dataflow tasks, resembling the load, compute, store design paradigm, but without the drawback of manual algorithm refactoring. Experimental results collected from an FPGA board show that our cache speeds up the execution of a variety of benchmarks by up to 60 times compared to the out-of-the-box solution provided by HLS, requiring very limited optimization effort. Our caches are not meant to compete with manually optimized implementations quality of results (QoR), but rather to significantly save design effort, in exchange for some QoR, to make the HLS flow a bit more software-like, allowing the designer to focus on algorithmic optimizations, rather than on explicit memory management. Moreover, caching could be the only feasible memory optimization for algorithms with data-dependent or irregular memory access patterns, but with good data locality.

21 March 2025 in ACM Transactions on Reconfigurable Technology and Systems
Verified
Journal Article
Open access
GBGiovanni Brignone
Giovanni Brignone
RBRoberto Bosio
Roberto Bosio
FOFabrizio Ottati
Fabrizio Ottati
CSClaudio Sansoè
Claudio Sansoè
Luciano LavagnoLuciano Lavagno
Luciano Lavagno
High-level synthesis (HLS) aims at democratizing custom hardware acceleration with highly abstracted software-like descriptions. However, efficient accelerators still require substantial low-level hardware optimizations, defeating the HLS intent. In the context of field-programmable gate arrays, digital signal processors (DSPs) are a crucial resource that typically requires a significant optimization effort for its efficient utilization, especially when used for sub-word vectorization. This work proposes SILVIA, an open-source LLVM transformation pass that automatically identifies superword-level parallelism within an HLS design and exploits it by packing multiple operations, such as additions, multiplications, and multiply-and-adds, into a single DSP. SILVIA is integrated in the flow of the commercial AMD Vitis HLS tool and proves its effectiveness by packing multiple operations on the DSPs without any manual source-code modifications on several diverse state-of-the-art HLS designs such as convolutional neural networks and basic linear algebra subprograms accelerators, reducing the DSP utilization for additions by 70% and for multiplications and multiply-and-adds by 50% on average.

14 April 2025 in IEEE Access
Verified
Journal Article
Open access
RBRoberto Bosio
Roberto Bosio
GBGiovanni Brignone
Giovanni Brignone
TUTeodoro Urso
Teodoro Urso
Mihai T. LazarescuMihai T. Lazarescu
Mihai T. Lazarescu
Luciano LavagnoLuciano Lavagno
Luciano Lavagno
Paolo PasiniPaolo Pasini
Paolo Pasini
Subgraph matching is a significant problem in several fields, including like social network analysis, chemical compound search, and fraud detection. While current solutions using CPU, graphics processing units (GPUs), and data center field-programmable gate arrays (FPGAs) deliver high performance, they consume significant power, making them unsuitable for resource-constrained edge environments. This articlea introduces a novel low-power FPGA accelerator that enables efficient subgraph matching using a small FPGA-based accelerator through three main innovations: a flexible two-level hash table architecture that minimizes memory accesses, a caching mechanism that reduces on-chip memory requirements, and a dynamic first-in first-out (FIFO) queue that efficiently buffers partial results. Experimental results on five real-world graphs show that the accelerator reduces energy consumption by an average of 75× compared to state-of-the-art CPU solutions, and 107× compared to GPU solutions, demonstrating that complex graph operations can be performed efficiently using even a small accelerator implemented on a reconfigurable platform. aA preliminary version of this work was presented as extended abstract in [1]