Listar Arquitectura de Computadores - (AC) por título
Mostrando ítems 44-63 de 183
-
DNA Sequences Alignment in Multi-GPUs: Energy Payoff on Speculative Executions
Pérez-Serrano, Jesús; Sandes, Edans; Melo, Alba; Ujaldon-Martínez, Manuel
(2017-05-30)
We present a performance per watt analysis of CUDAlign 4.0, a parallel strategy to obtain the optimal alignment of huge DNA se- quences in multi-GPU platforms using the exact Smith-Waterman method. Speed-up factors and ... -
Edge AI Architectures for a Privacy-Preserving IoT Era
(2022-07-20)The Internet of Things (IoT) has been hailed as the next frontier of innovation where our everyday objects are connected in ways that improve our lives and transform industries, in particular healthcare. In this talk, Prof. ... -
Efficient floating-point givens rotation unit
Hormigo-Aguilar, Javier; Muñoz, Sergio (2020-10-23)
High-throughput QR decomposition is a key operation in many advanced signal processing and communication applications. For some of these applications, using floating-point computation is becoming almost compulsory. However, ... -
Efficient Floating-Point Representation for Balanced Codes for FPGA Devices
Villalba-Moreno, Julio; Hormigo-Aguilar, Javier; Corbera-Peña, Francisco Javier; González, Mario; López-Zapata, Emilio (2013-10-30)
We propose a floating–point representation to deal efficiently with arithmetic operations in codes with a balanced number of additions and multiplications for FPGA devices. The variable shift operation is very slow in ... -
Efficient heterogeneous matrix profile on a CPU + High Performance FPGA with integrated HBM
Romero Moreno, José Carlos; González-Navarro, María Ángeles; Vilches Reina, Antonio; Corbera-Peña, Francisco Javier (Elsevier, 2021-12)
In this work, we study the problem of efficiently executing a state-of-the-art time series algorithm class – SCAMP – on a heterogeneous platform comprised of CPU + High Performance FPGA with integrated HBM (High Bandwidth ... -
Efficient OpenCL-based concurrent tasks offloading on accelerators
Lázaro-Muñoz, Antonio J.; González-Linares, José María; Gómez-Luna, Juan; Guil-Mata, Nicolás (Procedia Computer Science, 2017)
Current heterogeneous platforms with CPUs and accelerators have the ability to launch several independent tasks simultaneously, in order to exploit concurrency among them. These tasks typically consist of data transfer ... -
Enabling Fast and Energy-Efficient FM-index Exact Matching using Processing-Near-Memory
Herruzo-Ruiz, José Manuel; Fernández, Iván; González-Navarro, Sonia; Óscar, Plata (Springer, 2021-09)
Memory bandwidth and latency constitutes a major performance bottleneck for many data-intensive applications. While high-locality access patterns take advantage of the deep cache hierarchies available in modern processors, ... -
End-to-end Incremental Learning
Castro Payán, Francisco Manuel; Marín-Jiménez, Manuel J.; Guil-Mata, Nicolás; Schmid, Cordelia; Alahari, Karteek (2018-07-06)
Although deep learning approaches have stood out in recent years due to their state-of-the-art results, they continue to suffer from (catastrophic forgetting), a dramatic decrease in overall performance when training with ... -
Energy Efficiency of Software Transactional Memory in a Heterogeneous Architecture
Villegas Fernández, Emilio; Villegas Fernández, Alejandro; González-Navarro, María Ángeles; Asenjo-Plaza, Rafael; Ukidave, Yash; Plata-González, Óscar Guillermo[et al.] (2016-09-07)
Hardware vendors make an important effort creating low-power CPUs that keep battery duration and durability above acceptable levels. In order to achieve this goal and provide good performance-energy for a wide variety of ... -
Energy-based tuning of convolution neural networks on multi-GPUs
Castro Payán, Francisco Manuel; Guil-Mata, Nicolás; Marín Jiménez, Manuel Jesús; Pérez-Serrano, Jesús; Ujaldon-Martínez, Manuel (Wiley, 2019-11)
Deep Learning (DL) applications are gaining momentum in the realm of Artificial Intelligence, particularly after GPUs have demonstrated remarkable skills for accelerating their challenging computational requirements. Within ... -
Energy-based tuning of metaheuristics for molecular docking on multi-GPUs
Pérez-Serrano, Jesús; Imbernón, Baldomero; Cecilia, José María; Ujaldon-Martínez, Manuel (Wiley, 2018-09)
Virtual Screening (VS) methods simulate molecular interactions in silico to look for the best chemical compound that interacts with a given molecular target. VS is becoming increasingly popular to accelerate the drug ... -
Entropy-based High Performance Computation of Boolean SNP-SNP Interactions Using GPUs
Riveros, Carlos; Ujaldon-Martínez, Manuel; Moscato, Pablo (2014-05-02)
It is being increasingly accepted that traditional statistical Single Nucleotide Polymorphism (SNP) analysis of Genome-Wide Association Studies (GWAS) reveals just a small part of the heritability in complex diseases. ... -
Evaluación del consumo energético de la memoria transaccional en procesadores heterogéneos
Villegas Fernández, Emilio; Villegas Fernández, Alejandro; González-Navarro, María Ángeles; Asenjo-Plaza, Rafael; Plata-González, Óscar Guillermo (2016)
Actualmente existe una enorme cantidad de dispositivos y sistemas, como ordenadores portátiles y teléfonos móviles, que dependen de una batería para su funcionamiento. Como consecuencia, el hardware que incorporan debe ser ... -
Evaluación Realista de Modelos de Aprendizaje Profundo para Imágenes Hiperespectrales
(2019-12-13)Con los recientes avances realizados en el campo de Observación de la Tierra (EO), el uso de información de detección remota capturada por sensores disponibles (ubicados en plataformas aéreas y / o satelitales) ha adquirido ... -
Evaluation of CNN architectures for gait recognition based on optical flow maps
Castro, Francisco M.; Marín-Jiménez, Manuel J.; Guil-Mata, Nicolás; López-Tapia, S.; Pérez de la Blanca, N. (2017)
This work targets people identification in video based on the way they walk (\ie gait) by using deep learning architectures. We explore the use of convolutional neural networks (CNN) for learning high-level descriptors ... -
Experiments with Active-Set LP Algorithms Allowing Basis Deficiency
Hendrix, Eligius María Theodorus; Guerrero-García, Pablo (IOAP-MDPI, 2022-12-23)
n interesting question for linear programming (LP) algorithms is how to deal with solutions in which the number of nonzero variables is less than the number of rows of the matrix in standard form. An approach is that of ... -
Exploiting Vector Extensions to Accelerate Time Series Analysis.
Quislant-del-Barrio, Ricardo; Fernández-Vega, Iván; Serralvo, Eduardo; Gutiérrez-Carrasco, Eladio Damián; Plata-González, Óscar Guillermo (SARTECO, 2022-09-21)
Time series analysis is an important research topic and a key step in monitoring and predicting events in many fields. Recently, the Matrix Profile method, and particularly two of its Euclidean-distance-based implementations ... -
Exploring multiprocessor approaches to time series analysis
Quislant-del-Barrio, Ricardo; Gutiérrez-Carrasco, Eladio Damián; Plata-González, Óscar Guillermo (Elsevier, 2024-02-08)
A time series is a chronologically ordered set of samples of a real-valued variable that can have millions of observations. Time series analysis seeks extracting models in a large variety of domains [31] such as epidemiology, ... -
Explotando el nuevo módulo OpenCL de Intel TBB
Romero, José Carlos; Villegas Fernández, Alejandro; González-Navarro, María Ángeles; Rodríguez-Moreno, Andrés; Asenjo-Plaza, Rafael (2018-07-20)
Este artículo tiene como objetivo contribuir al desarrollo de la programación paralela trabajando en una de las herramientas desarrolladas por Intel: Intel Threading Building Blocks (Intel TBB). Hemos implementado una ... -
Expressing Heterogeneous Parallelism in C++ with Threading Building Blocks
Reinders, James; Voss, Michael J.; Reble, Pablo; Asenjo-Plaza, Rafael (2017-12-18)
Due to energy constraints, high performance computing platforms are becoming increasingly heterogeneous, achieving greater performance per watt through the use of hardware that is tuned to specific computational kernels ...