Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 19
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

A performance comparison of CUDA remote GPU virtualization frameworks

  • Reaño González, Carlos
  • Silla, Federico|||0000-0002-6435-1200
Using GPUs reduces execution time of many applications but increases acquisition cost and power consumption. Furthermore, GPUs usually attain a relatively low utilization. In this context, remote GPU virtualization solutions were recently created to overcome the drawbacks of using GPUs. Currently, many different remote GPU virtualization frameworks exist, all of them presenting very different characteristics. These differences among them may lead to differences in performance. In this work we present a performance comparison among the only three CUDA remote GPU virtualization frameworks publicly available at no cost. Results show that performance greatly depends on the exact framework used, being the rCUDA virtualization solution the one that stands out among them. Furthermore, rCUDA doubles performance over CUDA for pageable memory copies., This work was funded by the Generalitat Valenciana under Grant PROMETEOII/2013/009 of the PROMETEO program phase II. Authors are also grateful for the generous support provided by Mellanox Technologies
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Real-time functionalization and biosensing in subwavelength grating bimodal waveguides

  • Torrijos-Morán, Luis
  • Martinez-Perez, Paula
  • García-Rupérez, Jaime|||0000-0002-2036-0084
[EN] In this paper, we present for the first time experimental results of biosensing processes by using subwavelength grating bimodal waveguides as a refractive index sensors. We demonstrate the detection of 10ppm concentration of Bovine Serum Albumin (BSA) with a 190pm wavelength shift after a functionalization of the sensor surface with a layer of protein A/G in a continuous flowing system. Real-time data of each step is obtained, validating these kind of novel and highly sensitive SWG sensors for biosensing purposes., The authors acknowledge the funding from the Generalitat Valencia through project AVANTI/2019/123 and grant ACIF/2019/009.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

1D Periodic Corrugated Waveguides for Real-Time Detection of Increasing Concentrations of Thrombin

  • Martinez-Perez, Paula
  • Torrijos-Morán, Luis
  • Gómez-Gómez, María Isabel|||0000-0001-7548-209X
  • Ivanova-Angelova, Todora|||0000-0002-8150-4133
  • Griol Barres, Amadeu
  • Hurtado Montañés, Juan|||0000-0002-4870-7641
  • Bellieres, Laurent Christophe
  • García-Rupérez, Jaime|||0000-0002-2036-0084
[EN] An aptasensor based on periodic 1D corrugated waveguides is demonstrated to be suitable for the specific and continuous detection of thrombin at nM concentrations in real time with an estimated detection limit of 33.5 pM., This research was supported by a co-financed action by the European Union through the operational program of the European Regional Development Fund (FEDER) of the Valencian Community 2014-2020, the Generalitat Valenciana through the PROMETEO project AVANTI/2019/123 and the grant ACIF/2019/009, and by Universitat Politècnica de València through grants PAID-01-17 and the project OCUSENSOR.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

A Heuristic optimization approach to solve berth allocation problem

  • Burgos-Simon, Clara|||0000-0001-6385-4263
  • Cortés, J.-C.|||0000-0002-6528-2155
  • Martínez-Rodríguez, David|||0000-0002-2093-7840
  • Villanueva Micó, Rafael Jacinto|||0000-0002-0131-0532
This work has been partially supported by the Ministerio de Economía y Competitividad grant MTM2017-89664-P and by the Ministerio de Ciencia, Innovacion y Universidades (Retos Colaboracion 2017) grant RTC-2017-6566-4, VALKNUT.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Subwavelength grating bimodal waveguide for refractive index sensing

  • Torrijos-Morán, Luis
  • García-Rupérez, Jaime|||0000-0002-2036-0084
[EN] Periodic subwavelength structures supporting two TE modes are presented as high performance sensors with bulk and surface sensitivities of 1375.5nm/RIU and 6.138nm/nm, respectively. A complete theoretical study is provided by numerical expressions and FDTD simulations.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Increasing the Performance of Data Centers by Combining Remote GPU Virtualization with Slurm

  • Iserte, Sergio
  • Reaño, Carlos
  • Prades, Javier|||0000-0003-3349-2200
  • Silla, Federico|||0000-0002-6435-1200
[EN] The use of Graphics Processing Units (GPUs) presents several side effects, such as increased acquisition costs as well as larger space requirements. Furthermore, GPUs require a non-negligible amount of energy even while idle. Additionally, GPU utilization is usually low for most applications. Using the virtual GPUs provided by the remote GPU virtualization mechanism may address the concerns associated with the use of these devices. However, in the same way as workload managers map GPU resources to applications, virtual GPUs should also be scheduled before job execution. Nevertheless, current workload managers are not able to deal with virtual GPUs. In this paper we analyze the performance attained by a cluster using the rCUDA remote GPU virtualization middleware and a modified version of the Slurm workload manager, which is now able to map remote virtual GPUs to jobs. Results show that cluster throughput is doubled at the same time that total energy consumption is reduced up to 40%. GPU utilization is also increased., This work was funded by Generalitat Valenciana under Grant PROMETEOII/2013/009 of the PROMETEO program phase II. The author from Universidad Jaume I was supported by project TIN2014-53495-R from MINECO and FEDER. The authors are grateful for the generous support provided by Mellanox Technologies and the equipment donated by NVIDIA Corporation.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

High performance photonic biosensors based on periodic configurations

  • García-Rupérez, Jaime|||0000-0002-2036-0084
  • Gómez-Gómez, María Isabel|||0000-0001-7548-209X
  • Ponce-Alcántara, Salvador|||0000-0002-7479-8129
  • Torrijos-Morán, Luis
  • Martinez-Perez, Paula
[EN] Periodic photonic configurations as photonic crystals (PhCs) and subwavelength grating (SWG) waveguides are gaining a renewed interest for the development of biosensing structures. By performing a proper design, these periodic configurations allow a significant sensitivity increase while keeping a compact footprint, what is achieved by exploiting concepts such as the slow-wave effect, the increase of the light-matter overlap or the interference of dispersion engineered modes., These results were achieved thanks to the funding received from the European Union (ICT-644242-SAPHELY, PHC634013-PHOCNOSIS and the operational program of the European Regional Development Fund (FEDER) of the Valencia Regional Government 2014¿2020), the Spanish Government (TEC2015-63838-C3-1-R-OPTONANOSENS and PID2019-106965RB-C21-PHOLOW), the Generalitat Valenciana (AVANTI/2019/123, ACIF/2019/009 and PPC/2020/037), the Universitat Politècnica de València (PAID-01-17, PAID-01-18 and OCUSENSOR).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Epidemiological modelling of COVID-19 in Granada, Spain: Analysis of future scenarios

  • Garrido, José-Manuel
  • Rodríguez-Serrano, Fernando
  • Martínez-Rodríguez, David|||0000-0002-2093-7840
  • San Julián-Garcés, Raúl|||0000-0003-3052-3200
  • Villanueva Micó, Rafael Jacinto|||0000-0002-0131-0532
The Spanish Ministerio de Economia, Industria y Competitividad (MINECO), the Agencia Estatal de Investigacion (AEI) and Fondo Europeo de Desarrollo Regional (FEDER UE) grant MTM2017-89664-P; the European Union through the Operational Program of the [European Regional Development Fund (ERDF) / European Social Fund (ESF)] of the Valencian Community 2014-2020. Files: GJIDI/2018/A/010 and GJIDI/2018/A/009; the Ramon Areces Foundation, Madrid, Spain (CIVP18A3920).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

InfiniBand verbs optimizations for remote GPU virtualization

  • Reaño González, Carlos
  • Silla, Federico|||0000-0002-6435-1200
The use of InfiniBand networks to interconnect high performance computing clusters has considerably increased during the last years. So much so that the majority of the supercomputers included in the TOP500 list either use Ethernet or InfiniBand interconnects. Regarding the latter, due to the complexity of the InfiniBand programming API (i.e., InfiniBand Verbs) and the lack of documentation, there are not enough recent available studies explaining how to optimize applications to get the maximum performance from this fabric. In this paper we expose two different optimizations to be used when developing applications using InfiniBand Verbs, each providing an average bandwidth improvement of 3.68% and 217.14%, respectively. In addition, we show that when combining both optimizations, the average bandwidth gain is 43.29%. This bandwidth increment is key for remote GPU virtualization frameworks. Actually, this noticeable gain translates into a reduction of up to 35% in execution time of applications using remote GPU virtualization frameworks., This work was funded by the Generalitat Valenciana under Grant PROMETEOII/2013/009 of the PROMETEO program phase II. Authors are also grateful for the generous support provided by Mellanox Technologies
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Multifocal acoustic holograms for deep-brain neuromodulation and BBB opening

  • Andrés, Diana
  • Jiménez-Gambín, Sergio
  • Jimenez, Noe|||0000-0002-6539-670X
  • Camarena Femenia, Francisco|||0000-0002-6713-1414
[EN] Single-element focused ultrasound devices have demonstrated its ability to non-invasively open the Blood Brain Barrier. However, skull irregularities and absorption produce strong aberrations in the ultrasound focus. Recently, 3D-printed acoustic holograms were used to compensate aberrations and to produce a sharp focus. In addition, using this technology the geometry of the ultrasonic focus can be matched to the shape of the brain structure of interest. In this work, we experimentally and numerically report 3D-printed acoustic holograms for bilateral focusing through an ex-vivo human skull. Using holograms, we target different cerebral nuclei and the ultrasonic focuses are adapted to the target volumes, minimizing the acoustic field outside. Simultaneously, skull aberrations are corrected using phased-conjugation methods. The holographic surfaces were manufactured for a 100-mm aperture and 140-mm focal focused transducer at 500 kHz to focus through an ex-vivo human skull. Numerical and experimental results agree targeting several deep-brain nuclei: left-and-right putamen, caudate nuclei, and hippocampi, showing the potential of this low-cost technology to optimize BBBO or neuromodulation treatments., This research has been supported by the Spanish Ministry of Science, Innovation and Universities through grant Juan de la Cierva - Incorporacion (IJC2018-037897-I) and PID2019-111436RBC22, by the Agencia Valenciana de la Innovacio through grants INNVAL10/19/016 and INNCON/2020/009. Action cofinanced by the European Union through the Programa Operativo del Fondo Europeo de Desarrollo Regional (FEDER) of the Comunitat Valenciana 2014-2020 (IDIFEDER/2018/022).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Bimodal waveguide sensors enabled by subwavelength grating structures

  • Torrijos-Morán, Luis
  • García-Rupérez, Jaime|||0000-0002-2036-0084
  • Griol Barres, Amadeu
[EN] A subwavelength grating sensor based on a bimodal waveguide configuration is presented for continuous in-flow measurements of refractive index variations. An experimental bulk sensitivity of 1350nm/RIU and a limit of detection of 2x10-5RIU is obtained in a single-channel refractive index sensor.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Remote GPU Virtualization: Is It Useful?

  • Silla, Federico|||0000-0002-6435-1200
  • Prades, Javier|||0000-0003-3349-2200
  • Iserte, Sergio
  • Reaño, Carlos
[EN] Graphics Processing Units (GPUs) are currently used in many computing facilities. However, GPUs present several side effects, such as increased acquisition costs as well as larger space requirements. Also, GPUs still require some amount of energy while idle and their utilization is usually low. In a similar way to virtual machines, using virtual GPUs may address the mentioned concerns. In this regard, remote GPU virtualization allows to share the GPUs present in the computing facility among the nodes of the cluster. This would increase overall GPU utilization, thus reducing the negative impact of the increased costs mentioned before. Reducing the amount of GPUs installed in the cluster could also be possible. In this paper we explore some of the benefits that remote GPU virtualization brings to clusters. For instance, this mechanism allows an application to use all the GPUs present in a cluster. Another benefit of this technique is that cluster throughput, measured as jobs completed per time unit, is doubled when this technique is used. Furthermore, in addition to increasing overall GPU utilization, total energy consumption is reduced up to 40%. This may be key in the context of exascale computing facilities, which present an important energy constraint., This work was funded by Generalitat Valenciana under Grant PROMETEOII/2013/009 of the PROMETEO program phase II. The author from Universitat Jaume I was supported by project TIN2014-53495-R from MINECO and FEDER. The authors are grateful for the generous support provided by Mellanox Technologies and the equipment donated by NVIDIA Corporation.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

On the Execution of Computationally Intensive CPU-based Libraries on Remote Accelerators to Increase Performance: Early Experience with the OpenBLAS and FFTW Libraries

  • Mislata Valero, Santiago
  • Silla, Federico|||0000-0002-6435-1200
Virtualization techniques have shown to report benefits to data centers and other computing facilities. In this regard, virtual machines not only allow reducing the size of the computing infrastructure while increasing overall resource utilization but virtualizing individual components of computers may also provide significant benefits. This is the case, for example, for the remote GPU virtualization technique, implemented in several frameworks during the last years. In this paper we present an initial implementation of a new middleware for the remote virtualization of another component of computers: the CPU itself. Our proposal uses remote accelerators to perform computations that were initially intended to be carried out in the local CPUs, doing so transparently to the application and without having to modify its source code. By making use of the OpenBLAS and FFTW libraries as case studies to show the performance gains of our proposal, we carry out a performance evaluation targeting several system configurations comprising Xeon processors as well as Ethernet and InfiniBand QDR, FDR, and EDR network adapters in addition to NVIDIA Tesla K40 GPUs. Results not only demonstrate that the new middleware is feasible, but they also show that mathematical libraries may experience a significant speed up, despite of having to move data forth and back to/from remote servers., This work was funded by the Generalitat Valenciana under Grant PROMETEOII/2013/009 of the PROMETEO program phase II.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Acceleration-as-a-Service: Exploiting Virtualised GPUs for a Financial Application

  • Varghese, Blesson
  • Reaño, Carlos
  • Prades, Javier|||0000-0003-3349-2200
  • Silla, Federico|||0000-0002-6435-1200
[EN] How can GPU acceleration be obtained as a service in a cluster? This question has become increasingly significant due to the inefficiency of installing GPUs on all nodes of a cluster. The research reported in this paper is motivated to address the above question by employing rCUDA (remote CUDA), a framework that facilitates Acceleration-as-a-Service (AaaS), such that the nodes of a cluster can request the acceleration of a set of remote GPUs on demand. The rCUDA framework exploits virtualisation and ensures that multiple nodes can share the same GPU. In this paper we test the feasibility of the rCUDA framework on a real-world application employed in the financial risk industry that can benefit from AaaS in the production setting. The results confirm the feasibility of rCUDA and highlight that rCUDA achieves similar performance compared to CUDA, provides consistent results, and more importantly, allows for a single application to benefit from all the GPUs available in the cluster without loosing efficiency., This research was supported by an NVIDIA award, the EPSRC EP/K015745/1 grant, and the Generalitat Valenciana PROMETEOII/2013/009 grant of the PROMETEO program phase II. The authors acknowledge the generous support of Mellanox Technologies.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Network computational model to estimate the effectiveness of the influenza vaccine a posteriori

  • Martínez-Rodríguez, David|||0000-0002-2093-7840
  • San Julián-Garcés, Raúl|||0000-0003-3052-3200
  • López-Navarro, Elena|||0000-0003-0164-9007
  • Villanueva Micó, Rafael Jacinto|||0000-0002-0131-0532
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

A new elastographic technique using acoustic vortices

  • Jimenez, Noe|||0000-0002-6539-670X
  • Benlloch Baviera, Jose María|||0000-0001-6073-1436
  • Camarena Femenia, Francisco|||0000-0002-6713-1414
[EN] We present a novel method of elastography based on acoustic vortices to transfer angular momentum to tissue in addition to linear momentum. Focused vortex beams can push and twist tissue, and the rotation direction of the applied torque can be dynamically controlled by the modulation of the topological charge of the vortex. The technique results in a robust excitation of shear waves with quasi-omnidirectional radiation pattern and arbitrary waveform, which may have a great impact in imaging performance for elastography., This research has been supported by the Spanish Ministry of Science, Innovation and Universities through grants Juan de la Cierva - Incorporacion IJC2018-037897-I and PID2019-111436RBC22, and by the Agencia Valenciana de la Innovacio through grants INNVA1/2020/92 and INNCON/2020/009. Action cofinanced by the European Union through the Programa Operativo del Fondo Europeo de Desarrollo Regional (FEDER) of the Comunitat Valenciana 2014-2020 (IDIFEDER/2018/022).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

CUDA acceleration for Xen virtual machines in infiniband clusters with rCUDA

  • Prades, Javier|||0000-0003-3349-2200
  • Silla, Federico|||0000-0002-6435-1200
  • Reaño, Carlos
[EN] Many data centers currently use virtual machines (VMs) to achieve a more efficient usage of hardware resources. However, current virtualization solutions, such as Xen, do not easily provide graphics processing unit (GPU) accelerators to applications running in the virtualized domain with the flexibility usually required in data centers (i.e., managing virtual GPU instances and concurrently sharing them among several VMs). Remote GPU virtualization frameworks such as the rCUDA solution may address this problem. In this work we analyze the use of the rCUDA framework to accelerate scientific applications running inside Xen VMs. Results show that the use of the rCUDA framework is a feasible approach, featuring a very low overhead if an InfiniBand fabric is already present in the cluster., This work was funded by the Generalitat Valenciana under Grant PROMETEOII/2013/009 of the PROMETEO program phase II. Authors are also grateful for the equipment donated by Mellanox Technologies and NVIDIA Corporation.
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Contribución de congreso (ConferenceOutput)

Posicionamiento en Interiores de Alta Precisión Utilizando Tecnología Infrarroja

  • Rodríguez Navarro, David|||0000-0002-7916-4221
  • Lázaro Galilea, José Luis|||0000-0001-5048-7134
  • Llana Calvo, Álvaro de la|||0000-0002-8889-0452
  • Gardel Vicente, Alfredo|||0000-0001-7887-4689
  • Bravo Muñoz, Ignacio|||0000-0002-6964-0036
  • Espinosa Zapata, Felipe|||0000-0003-1588-0947
SEMINARIO ANUAL AUTOMATICA ELECTRONICA INDUSTRIAL E INSTRUMENTACION - SAAEI, , 03/07/2019-05/07/2019, Cordoba, España, El trabajo presenta los resultados del sistema de posicionamiento en interiores basado en infrarrojo desarrollado por nuestro grupo de investigación. Dado que uno de los principales problemas que se presentan en todas las tecnologías para posicionamiento en interiores es el de los multicaminos (MP) debidos a reflexiones en objetos y paredes, en una primera fase del trabajo presentamos los estudios realizados para analizar cómo afectan en función de las diferentes técnicas de medida utilizadas y cómo se pueden mitigar. A continuación, se presenta nuestra propuesta tecnológica para el desarrollo de un sistema de posicionamiento basado en infrarrojos (IR), seguido de los resultados obtenidos. Como se muestra, los errores de posicionamiento con un solo detector, trabajando a la vez en, Universidad de Alcalá, Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Contribución de congreso (ConferenceOutput)

Sistema de posicionamiento en interiores basado en iluminación LED y sensor PSD

  • Llana Calvo, Álvaro de la|||0000-0002-8889-0452
  • Lázaro Galilea, José Luis|||0000-0001-5048-7134
  • Gardel Vicente, Alfredo|||0000-0001-7887-4689
  • Rodríguez Navarro, David|||0000-0002-7916-4221
  • Bravo Muñoz, Ignacio|||0000-0002-6964-0036
  • Espinosa Zapata, Felipe|||0000-0003-1588-0947
26º Seminario Anual de Automática Electrónica Industrial e Instrumentación (SAAEI 2019), 03/07/2019-05/07/2019, Cordoba, España., En este trabajo proponemos un sistema y un método para posicionamiento en interiores basado en la propia iluminación del entorno con dispositivos LEDs. La propuesta se basa en la modulación de la señal visible emitida, en frecuencias entre los 50 y 200 kHz y su detección sobre un fotodiodo PSD (Position Sensitive Detector). Se presenta una primera aproximación del modelo y de su implementación en los que se asume que el detector se desplaza sobre un plano (en este caso a distancias entre 1 y 3 metros de la fuente de luz) y se detecta su localización a partir de un punto de luz. La técnica para llevar a cabo el posicionamiento se basa en el ángulo de llegada de la señal luminosa (AoA). En el trabajo se han utilizado algoritmos genéticos para determinar los parámetros del modelo del detector (compuesto por un PSD y lente acoplada) y posteriormente se han realizado pruebas para determinar la precisión en el posicionamiento a 3 metros de distancia y sobre un área de cobertura de 3x3 m2. En el apartado de resultados se refleja el error medio en la detección de la posición. Como se puede observar, con bajo efecto de los multicaminos (MP) el error en posicionamiento está por debajo de 1 cm en el 80% de los puntos. Con la presencia de efectos fuertes de multicaminos los errores máximos en el posicionamiento no superan los 4,5 cm., Universidad de Alcalá, Agencia Estatal de Investigación