Luis Gerardo León-Vega
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Luis Gerardo León-Vega has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Jorge Castro Godinez
Consejo Nacional de Rectores (CONARE)
4 shared publications - Alfonso Chacon-Rodriguez
Department of Electronics Engineering, Instituto Tecnológico de Costa Rica, Cartago 30101, Costa Rica
2 shared publications - Carlos Salazar-García
Tecnológico de Costa Rica
2 shared publications - Kaleb Alfaro-Badilla
Tecnológico de Costa Rica
2 shared publications - Niccolò Tosato
Università degli Studi di Trieste
2 shared publications - Stefano Cozzini
Area Science Park, Trieste, Italy
2 shared publications
Latest Publications
- 23 June 2026
- Published by Pleiades Publishing Ltd in Programming and Computer Software
- Vol. 52 (3) , 239-255
Integrated Software and Thermal Imaging Framework to Evaluate Passive PV Cooling Systems in Intelligent Urban Environments
- 23 June 2026
- Published by Pleiades Publishing Ltd in Programming and Computer Software
- Vol. 52 (3) , 239-255
Integrated Software and Thermal Imaging Framework to Evaluate Passive PV Cooling Systems in Intelligent Urban Environments
- 14 February 2025
- Published by Springer Nature
- p. 52-67
A Comprehensive Analysis of Process Energy Consumption on Multi-socket Systems with GPUs
- 14 February 2025
- Published by Springer Nature
- p. 52-67
A Comprehensive Analysis of Process Energy Consumption on Multi-socket Systems with GPUs
- 14 February 2025
- Published by Springer Nature
- p. 112-126
EfiMon: A Process Analyser for Granular Power Consumption Prediction
- 14 February 2025
- Published by Springer Nature
- p. 112-126
EfiMon: A Process Analyser for Granular Power Consumption Prediction
- 18 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
CYNQ: Speeding Up FPGA Applications with Simplicity
FPGAs for hardware acceleration face several issues in their adoption by the emerging AI embedded software commu-nity, mainly due to the difficulty of getting started in the field. To address these issues, FPGA vendors have tried to simplify and combine workflows to make them user-friendly. This has also impacted the runtime libraries used on Linux to allow user space applications to interact with FPGA programmable logic. However, this approach can result in suboptimal results, given the added overhead in single-purpose applications in an attempt to generalise the runtime libraries. This work presents CYNQ Runtime Library, a simplistic and API-agnostic C/C++ runtime library with the ease of PYNQ and the efficiency of C++ and XRT. In comparison to most popular runtime libraries for AMD FPGAs, our work outperforms PYNQ and XRT with more than 5.34 × speedup in execution latency and more than 1.20 × in large workloads than XRT, which poses a better scenario for high-performance single-purpose applications.
- 18 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
CYNQ: Speeding Up FPGA Applications with Simplicity
FPGAs for hardware acceleration face several issues in their adoption by the emerging AI embedded software commu-nity, mainly due to the difficulty of getting started in the field. To address these issues, FPGA vendors have tried to simplify and combine workflows to make them user-friendly. This has also impacted the runtime libraries used on Linux to allow user space applications to interact with FPGA programmable logic. However, this approach can result in suboptimal results, given the added overhead in single-purpose applications in an attempt to generalise the runtime libraries. This work presents CYNQ Runtime Library, a simplistic and API-agnostic C/C++ runtime library with the ease of PYNQ and the efficiency of C++ and XRT. In comparison to most popular runtime libraries for AMD FPGAs, our work outperforms PYNQ and XRT with more than 5.34 × speedup in execution latency and more than 1.20 × in large workloads than XRT, which poses a better scenario for high-performance single-purpose applications.
- 29 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
A User-Friendly Ecosystem for AI FPGA-Based Accelerators
The introduction of FPGAs in High-Performance Embedded Computing and Artificial Intelligence still faces chal-enges regarding the difficulty of getting started. It requires hardware knowledge, familiarity with multiple tooling, libraries and frameworks and long synthesis times. To encourage the usage of FPGAs, this work proposes an ecosystem that includes a library with a set of pre-built accelerators for common Digital Signal Processing and Artificial Intelligence workloads, an engine for runtime arbitrary-precision quantisation and an agnostic API, allowing the development of FPGA-accelerated user applications while abstracting the details about the FPGA design and implementation. Our approach is based on hardware reuse, introducing software resource management of a series of pre-built IP cores, allowing low-end FPGAs to be used as hardware accelerators and multiple applications to share resources. Our work is better than managed FPGA standalone applications with Vitis HLS-based quantisation, accelerating 1.22 x, thanks to our quantisation engine, which accelerates 5.12 x the quantisation and 13.30 x the de-quantisation, while keeping close the accelerator execution times.
- 29 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
A User-Friendly Ecosystem for AI FPGA-Based Accelerators
The introduction of FPGAs in High-Performance Embedded Computing and Artificial Intelligence still faces chal-enges regarding the difficulty of getting started. It requires hardware knowledge, familiarity with multiple tooling, libraries and frameworks and long synthesis times. To encourage the usage of FPGAs, this work proposes an ecosystem that includes a library with a set of pre-built accelerators for common Digital Signal Processing and Artificial Intelligence workloads, an engine for runtime arbitrary-precision quantisation and an agnostic API, allowing the development of FPGA-accelerated user applications while abstracting the details about the FPGA design and implementation. Our approach is based on hardware reuse, introducing software resource management of a series of pre-built IP cores, allowing low-end FPGAs to be used as hardware accelerators and multiple applications to share resources. Our work is better than managed FPGA standalone applications with Vitis HLS-based quantisation, accelerating 1.22 x, thanks to our quantisation engine, which accelerates 5.12 x the quantisation and 13.30 x the de-quantisation, while keeping close the accelerator execution times.
- Open Access
- 24 July 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 22 (4) , 1-27
Automatic Generation of Resource and Accuracy Configurable Processing Elements
Low-power consumption and scarce computational resources limit the computation at the edge. Besides, the approximate computing paradigm reports promising techniques for designing accelerators to deal with inherent limitations of the edge, and high-level synthesis with C++ opens the opportunity to use meta-programming for specialisable generic design. This work proposes a framework for automatically generating synthesis-time configurable processing elements (PEs) for matrix multiplication-addition (GEMMA) and convolution. To evaluate our work, we perform a design exploration after varying data bit-width, operand sizes, and kernel sizes. Our analyses include resource consumption scaling, clocks-to-solution, design efficiency, and error distribution, presenting a comprehensive view of how the parameters affect the properties of our generic implementations. The GEMMA presented a trade-off between granularity vs efficiency , where large PEs with short data widths are favoured by the design efficiency, achieving, theoretically, up to 75 GMAC/s on a Xilinx XC7Z020 @ 100 MHz with an efficiency of 27%. For design efficiency, we propose a figure of merit to evaluate operations per second and resource utilisation with respect to the maximum achievable by the FPGA. Regarding the convolution PEs, we implemented two algorithms: a window-based spatial convolution and Winograd. The former is the best in terms of performance with 150 GMAC/s, reaching up to 47% of efficiency. Winograd also outperformed numerically using a 3× 3 kernel filter, presenting a mean error of 11.01% in 4-bits operands with a PSNR=16.28 dB, compared to the spatial convolution with 38.2% of mean error and PSNR=5.89 dB. Finally, we discuss how the error is mostly dependent on the PE’s parameters. In the GEMMA, the error depends on the matrix size, causing limitations in the PE scaling but still applicable to accelerators. The PEs developed during this research will lead to further granular approximate accelerator research.
- 24 July 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 22 (4) , 1-27
Open AccessAutomatic Generation of Resource and Accuracy Configurable Processing Elements
Low-power consumption and scarce computational resources limit the computation at the edge. Besides, the approximate computing paradigm reports promising techniques for designing accelerators to deal with inherent limitations of the edge, and high-level synthesis with C++ opens the opportunity to use meta-programming for specialisable generic design. This work proposes a framework for automatically generating synthesis-time configurable processing elements (PEs) for matrix multiplication-addition (GEMMA) and convolution. To evaluate our work, we perform a design exploration after varying data bit-width, operand sizes, and kernel sizes. Our analyses include resource consumption scaling, clocks-to-solution, design efficiency, and error distribution, presenting a comprehensive view of how the parameters affect the properties of our generic implementations. The GEMMA presented a trade-off between granularity vs efficiency , where large PEs with short data widths are favoured by the design efficiency, achieving, theoretically, up to 75 GMAC/s on a Xilinx XC7Z020 @ 100 MHz with an efficiency of 27%. For design efficiency, we propose a figure of merit to evaluate operations per second and resource utilisation with respect to the maximum achievable by the FPGA. Regarding the convolution PEs, we implemented two algorithms: a window-based spatial convolution and Winograd. The former is the best in terms of performance with 150 GMAC/s, reaching up to 47% of efficiency. Winograd also outperformed numerically using a 3× 3 kernel filter, presenting a mean error of 11.01% in 4-bits operands with a PSNR=16.28 dB, compared to the spatial convolution with 38.2% of mean error and PSNR=5.89 dB. Finally, we discuss how the error is mostly dependent on the PE’s parameters. In the GEMMA, the error depends on the matrix size, causing limitations in the PE scaling but still applicable to accelerators. The PEs developed during this research will lead to further granular approximate accelerator research.
- Open Access
- 28 November 2022
- Published by Instituto Tecnologico de Costa Rica in Revista Tecnología en Marcha
- Vol. 35, ág. 39-45
Accelerating machine learning at the edge with approximate computing on FPGAs
La inferencia en algoritmos complejos de aprendizaje automático (ML, por sus siglas en inglés) en edge computing está tomando importancia para desvincular la funcionalidad de un sistema de la nube. Sin embargo, los modelos de ML incrementan en complejidad más rápido que los recursos de hardware disponibles. Esta investigación tiene como objetivo acelerar el aprendizaje por automático al delegar el cálculo computacional a FPGAs de baja gama y usar computación aproximada para optimizar el uso de los recursos aprovechando la naturaleza inexacta de los modelos de aprendizaje automático. En este artículo, proponemos un diseño de un elemento de procesamiento genérico de multiplicación-suma de matrices, parametrizado en el tipo de dato, tamaño de la matriz y ancho del dato. Evaluamos el comportamiento del consumo de recursos y del error mientras variamos el tamaño de la matriz y el ancho del dato dato un tipo de datos de punto fijo. Determinamos que el error escala con el tamaño de la matriz pero que puede ser compensado al incrementar el ancho del dato, representando un compromiso entre el ancho del dato y el tamaño de la matriz con respecto al error.
- 28 November 2022
- Published by Instituto Tecnologico de Costa Rica in Revista Tecnología en Marcha
- Vol. 35, ág. 39-45
Open AccessAccelerating machine learning at the edge with approximate computing on FPGAs
La inferencia en algoritmos complejos de aprendizaje automático (ML, por sus siglas en inglés) en edge computing está tomando importancia para desvincular la funcionalidad de un sistema de la nube. Sin embargo, los modelos de ML incrementan en complejidad más rápido que los recursos de hardware disponibles. Esta investigación tiene como objetivo acelerar el aprendizaje por automático al delegar el cálculo computacional a FPGAs de baja gama y usar computación aproximada para optimizar el uso de los recursos aprovechando la naturaleza inexacta de los modelos de aprendizaje automático. En este artículo, proponemos un diseño de un elemento de procesamiento genérico de multiplicación-suma de matrices, parametrizado en el tipo de dato, tamaño de la matriz y ancho del dato. Evaluamos el comportamiento del consumo de recursos y del error mientras variamos el tamaño de la matriz y el ancho del dato dato un tipo de datos de punto fijo. Determinamos que el error escala con el tamaño de la matriz pero que puede ser compensado al incrementar el ancho del dato, representando un compromiso entre el ancho del dato y el tamaño de la matriz con respecto al error.
- Open Access
- 28 November 2022
- Published by Instituto Tecnologico de Costa Rica in Revista Tecnología en Marcha
- Vol. 35, ág. 46-52
Design of a microkernelbased services manager for IoT with isolated processing
El desarrollo de proyectos relacionados con el Internet de las Cosas (IoT, por sus siglas en inglés) han generado un impacto significativo en el crecimiento de la industria del software. Si bien existen alternativas para su desarrollo, los usuarios se ven limitados por factores inherentes, tales como la dependencia de un proveedor, flexibilidad de desarrollo de software y seguridad. Este artículo muestra el diseño de una plataforma que facilita el despliegue y manejo de proyectos de IoT, solventando algunos de los retos actuales del mercado como los bloqueos de proveedor y limitaciones en las tecnologías de desarrollo.
- 28 November 2022
- Published by Instituto Tecnologico de Costa Rica in Revista Tecnología en Marcha
- Vol. 35, ág. 46-52
Open AccessDesign of a microkernelbased services manager for IoT with isolated processing
El desarrollo de proyectos relacionados con el Internet de las Cosas (IoT, por sus siglas en inglés) han generado un impacto significativo en el crecimiento de la industria del software. Si bien existen alternativas para su desarrollo, los usuarios se ven limitados por factores inherentes, tales como la dependencia de un proveedor, flexibilidad de desarrollo de software y seguridad. Este artículo muestra el diseño de una plataforma que facilita el despliegue y manejo de proyectos de IoT, solventando algunos de los retos actuales del mercado como los bloqueos de proveedor y limitaciones en las tecnologías de desarrollo.
- Open Access
- 31 January 2022
- Published by Universidad Nacional de Costa Rica in Uniciencia
- Vol. 36 (1) , 1-9
Measuring Traffic Dynamics at the Edge
This work aims to measure the impact of approximate computing on a case study of traffic dynamics metering on a System-on-Chip edge computing device. Firstly, the study proposes a baseline implementation of the metering system in C++. To analyze the application in detail, study profiled the baseline using a built-in instrumented profiler, presenting the overall performance of each of its parts. During the hotspot analysis, some parts had optimization opportunities exploitable by multi-threading and approximate computing techniques, particularly frame skipping, which is inspired by the loop perforation approximate technique. The first optimization employed was multi-threading, which led to a 1.32x speedup on the application without introducing errors in the metrics. Then, the meter was optimized by using frame skipping. This work demonstrated that adaptatively modifying the number of frames skipped improved the error in the final metrics compared to keeping it fixed. In terms of performance, the frame skipping brought the overall speed up to 1.76x. Approximate computing, in particular, frame skipping, managed to contribute up to 25% of the overall speedup, managing to accelerate the meter from 8.7 frames per second to 15 fps in the most critical case in exchange for some numerical error on the final metrics.
- 31 January 2022
- Published by Universidad Nacional de Costa Rica in Uniciencia
- Vol. 36 (1) , 1-9
Open AccessMeasuring Traffic Dynamics at the Edge
This work aims to measure the impact of approximate computing on a case study of traffic dynamics metering on a System-on-Chip edge computing device. Firstly, the study proposes a baseline implementation of the metering system in C++. To analyze the application in detail, study profiled the baseline using a built-in instrumented profiler, presenting the overall performance of each of its parts. During the hotspot analysis, some parts had optimization opportunities exploitable by multi-threading and approximate computing techniques, particularly frame skipping, which is inspired by the loop perforation approximate technique. The first optimization employed was multi-threading, which led to a 1.32x speedup on the application without introducing errors in the metrics. Then, the meter was optimized by using frame skipping. This work demonstrated that adaptatively modifying the number of frames skipped improved the error in the final metrics compared to keeping it fixed. In terms of performance, the frame skipping brought the overall speed up to 1.76x. Approximate computing, in particular, frame skipping, managed to contribute up to 25% of the overall speedup, managing to accelerate the meter from 8.7 frames per second to 15 fps in the most critical case in exchange for some numerical error on the final metrics.
- 12 February 2020
- Published by Springer Nature
- p. 49-62
Optimizing Big Data Network Transfers in FPGA SoC Clusters: TECBrain Case Study
- 12 February 2020
- Published by Springer Nature
- p. 49-62
Optimizing Big Data Network Transfers in FPGA SoC Clusters: TECBrain Case Study