Pasar al contenido principal

NUEVOS DESAFIOS EN COMPUTACION DE ALTAS PRESTACIONES: DESDE ARQUITECTURAS HASTA APLICACIONES

TIN2013-42148-P

•
Nombre agencia financiadora Ministerio de Economía y Competitividad
Acrónimo agencia financiadora MINECO
Programa Programa Estatal de Fomento de la Investigación Científica y Técnica de Excelencia
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D (Excelencia / Generación de Conocimiento)
Año convocatoria 2013
Plan Estatal 2013-2016 Plan Estatal de Investigación Científica y Técnica y de Innovación
Unidad de gestión Secretaría de Estado de Educación, Formación Profesional y Universidades
Centro beneficiario UNIVERSIDADE DA CORUÑA
Identificador persistente http://dx.doi.org/10.13039/501100003329

Resultados relacionados

Resultados totales (Incluyendo duplicados): 5
Encontrada(s) 1 página(s)

RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

Efficient Scan Operator Methods on a GPU

  • Pérez Diéguez, Adrián
  • Amor, Margarita
  • Doallo, Ramón
Presented at: 2014 IEEE 26th International Symposium on Computer Architecture and High Performance Computing, Paris, France, 22-24 October 2014, This version of the article has been accepted for publication, after peer review. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The Version of Record is available online at: https://doi.org/10.1109/SBAC-PAD.2014.23, [Abstract]: Current GPUs (Graphics Processing Units) offer high computational power at relatively low cost, nonetheless, this enhanced performance often comes at the expenses of flexibility and code complexity. Efficient GPU programming requires detailed knowledge on certain hardware aspects. The scan operator is an important building block for a wide range of algorithms. In this paper, we present a number of parallel scan methods based on the traditional cyclic reduction tridiagonal solver and the Ladner-Fischer parallel prefix adder. Futhermore, we analyze a set of new features introduced in the Kepler Nvidia architecture such as read-only data cache and shuffle instructions. Our methods provide an excellent performance in many cases, up to 48% improvement over the CUDA Data Parallel Primitives (CUDPP) library., This research has been supported by the Galician Government (Xunta deGalicia) under the Consolidation Program of Competitive Reference Groups, cofunded by FEDER funds of the EU (Ref. GRC2013/055); and by the Ministry of Economy and Competitiviness of Spain and FEDER funds of the EU (Project TIN2013–42148-P), Xunta de Galicia; GRC2013/055
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

GPU Accelerated Molecular Docking Simulation with Genetic Algorithms

  • Altuntas, Serkan
  • Bozkus, Zeki
  • Fraguela, Basilio B.
The final publication is available at Springer via https://doi.org/10.1007/978-3-319-31153-1_10, [Abstract] Receptor-Ligand Molecular Docking is a very computationally expensive process used to predict possible drug candidates for many diseases. A faster docking technique would help life scientists to discover better therapeutics with less effort and time. The requirement of long execution times may mean using a less accurate evaluation of drug candidates potentially increasing the number of false-positive solutions, which require expensive chemical and biological procedures to be discarded. Thus the development of fast and accurate enough docking algorithms greatly reduces wasted drug development resources, helping life scientists discover better therapeutics with less effort and time. In this article we present the GPU-based acceleration of our recently developed molecular docking code. We focus on offloading the most computationally intensive part of any docking simulation, which is the genetic algorithm, to accelerators, as it is very well suited to them. We show how the main functions of the genetic algorithm can be mapped to the GPU. The GPU-accelerated system achieves a speedup of around ~ 14x with respect to a single CPU core. This makes it very productive to use GPU for small molecule docking cases., Sekan Altuntaş and Zeki Bozkus are funded by the Scientific and Technological Research Council of Turkey (TUBITAK; 112E191). Basilio B. Fraguela is supported by the Ministry of Economy and Competitiveness of Spain and FEDER funds of the EU (ref. TIN2013-42148-P) and by the Galician Government under the Consolidation Program of Competitive Reference Groups (ref. GRC2013-055), Scientific and Technological Research Council of Turkey (TUBITAK); 112E19, Xunta de Galicia; GRC 2013-055
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

MREv: An Automatic MapReduce Evaluation Tool for Big Data Workloads

  • Veiga, Jorge
  • Expósito, Roberto R.
  • Taboada, Guillermo L.
  • Touriño, Juan
[Abstract]: The popularity of Big Data computing models like MapReduce has caused the emergence of many frameworks oriented to High Performance Computing (HPC) systems. The suitability of each one to a particular use case depends on its design and implementation, the underlying system resources and the type of application to be run. Therefore, the appropriate selection of one of these frameworks generally involves the execution of multiple experiments in order to assess their performance, scalability and resource efficiency. This work studies the main issues of this evaluation, proposing a new MapReduce Evaluator (MREv) tool which unifies the configuration of the frameworks, eases the task of collecting results and generates resource utilization statistics. Moreover, a practical use case is described, including examples of the experimental results provided by this tool. MREv is available to download at http://mrev.des.udc.es., This work was funded by the Ministry of Economy and Competitiviness of Spain (Ref. TIN2013- 42148-P), and by the Galician Government (Xunta de Galicia) under the Consolidation Program of Competitive Reference Groups (Ref. GRC2013/055), cofunded by FEDER funds of the EU., Xunta de Galicia; GRC2013/055
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

New Tridiagonal Systems Solvers on GPU Architectures

  • Pérez Diéguez, Adrián
  • Amor, Margarita
  • Doallo, Ramón
This version of the article has been accepted for publication, after peer review. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The Version of Record is available online at: https://doi.org/10.1109/HiPC.2015.17, [Abstract]: Modern GPUs (Graphics Processing Units) offer very high computing power at relatively low cost. Nevertheless, designing efficient algorithms for the GPUs usually requires additional time and effort, even for experienced programmers. On the other hand, tridiagonal systems solvers are an important building block for a wide range of applications. In this paper, we present a new tuning parallel proposal in order to generate new tridiagonal systems solvers. This proposal is based on the combination of a new reduction algorithm (Redundant Reduction-RR) with a tuning proposal to generate efficient parallel prefix algorithms on the GPU. Specifically, we present two new solvers combining RR with two GPU efficient parallel prefix patterns. The performance of the resulting proposals was analyzed using three different CUDA GPUs, obtaining an improvement of up to 20.5x over the CUSPARSE library and 28.9x over CUDPP., This research has been supported by the Galician Government (Xunta de Galicia) under the Consolidation Program of Competitive Reference Groups, cofunded by FEDER funds of the EU (Ref. GRC2013/055); and by the Ministry of Economy and Competitiveness of Spain and FEDER funds of the EU (Project TIN2013-42148-P). This work is also partially supported by EU under the COST Program Action IC1305: Network for Sustainable Ultrascale Computing (NESUS)., Xunta de Galicia; GRC2013/055, Presentado en: 2015 IEEE 22nd International Conference on High Performance Computing (HiPC), 16-19 December 2015
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Contribución de congreso (ConferenceOutput)

Implementing Parallel Differential Evolution on Spark

  • Teijeiro, Diego
  • Pardo, Xoán C.
  • González, Patricia
  • Banga, Julio R.
  • Doallo, Ramón
15 pages.-- European Conference on the Applications of Evolutionary Computation, EvoApplications 2016, Metaheuristics are gaining increased attention as an efficient way of solving hard global optimization problems. Differential Evolution (DE) is one of the most popular algorithms in that class. However, its application to realistic problems results in excessive computation times. Therefore, several parallel DE schemes have been proposed, most of them focused on traditional parallel programming interfaces and infrastructures. However, with the emergence of Cloud Computing, new programming models, like Spark, have appeared to suit with large-scale data processing on clouds. In this paper we investigate the applicability of Spark to develop parallel DE schemes to be executed in a distributed environment. Both the master-slave and the island-based DE schemes usually found in the literature have been implemented using Spark. The speedup and efficiency of all the implementations were evaluated on the Amazon Web Services (AWS) public cloud, concluding that the island-based solution is the best suited to the distributed nature of Spark. It achieves a good speedup versus the serial implementation, and shows a decent scalability when the number of nodes grows, This research received financial support from the Spanish Ministerio de Economia y Competitividad (and the FEDER) through the Project SYNBIOFACTORY (grant number DPI2014-55276-C5-2-R). It has been also supported by the Spanish Ministerio de Ciencia e Innovacion (and the FEDER) through the Project TIN2013-42148-P, and by the Galician Government (Xunta de Galicia) under the Consolidation Program of Competitive Research Units (Network Ref. R2014/041 and Project Ref. GRC2013/055) cofunded by FEDER funds of the EU, No