Pasar al contenido principal

ARQUITECTURAS, SISTEMAS Y HERRAMIENTAS PARA COMPUTACION DE ALTAS PRESTACIONES

TIN2010-16735

•
Nombre agencia financiadora Ministerio de Ciencia e Innovación
Acrónimo agencia financiadora MICINN
Programa Programa Nacional de Proyectos de Investigación Fundamental
Subprograma Subprograma de Proyectos de Investigación Fundamental no Orientada
Convocatoria Proyectos I+D
Año convocatoria 2010
Plan Estatal 2008-2011 Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica
Unidad de gestión Subdirección General de Proyectos de Investigación
Centro beneficiario UNIVERSIDADE DA CORUÑA
Identificador persistente http://dx.doi.org/10.13039/501100004837

Resultados relacionados

Resultados totales (Incluyendo duplicados): 8
Encontrada(s) 1 página(s)

RUC. Repositorio da Universidade da Coruña
Publicación Artículo científico (JournalArticle)

High-performance Monte Carlo radiosity on GPU based on scene partitioning

  • Sanjurjo Amado, José Rodrigo
  • Amor, Margarita
  • Bóo, Montserrat
  • Doallo, Ramón
This is the Accepted version of the published document. This version of the article: Sanjurjo, J. R., Amor, M., Bóo, M., & Doallo, R. (2012). ‘High-performance Monte Carlo radiosity on GPU based on scene partitioning’ has been accepted for publication in: Microprocessors and Microsystems, 36(2), 88-95. The Version of Record is available online at https://doi.org/10.1016/j.micpro.2011.05.004., [Abstract]: The recent interest in GPGPU, (General-Purpose computation on Graphics Processing Unit), has stimulated improvements in the programmability of the GPU. Although the utilization of new languages like OpenCL and CUDA facilitate GPU programming, different challenges have to be overcome to optimize the results of a direct implementation. Specifically, a straightforward implementation of the Monte Carlo radiosity algorithm on the GPU does not produce the expected performance. In this paper we develop different strategies to increase the performance of the implementation: utilization of an additional simplified version of the mesh to reduce the computational requirements, data partitioning of the scene to increase the data locality, and an efficient thread scheduling to exploit the characteristics of the GPU. Our approach increases the flexibility of previous solutions and the results show a significant improvement of the execution time., This work was supported by the Xunta de Galicia under projects INCITE08PXIB105161PR and 08TIC001206PR, the Ministry of Science and Innovation, cofunded by the FEDER funds of the European Union under the grant TIN2010-16735., Xunta de Galicia; INCITE08PXIB105161PR, Xunta de Galicia; 08TIC001206PR
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

SPLG: A Tuned Signal Processing Library for GPU Architectures

  • Lobeiras Blanco, Jacobo
  • Amor, Margarita
  • Doallo, Ramón
Presented at: 25th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2013, 23 through 26 October 2013, 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.2013.30, [Abstract]: In order to increase the efficiency of existing software many works are incorporating GPU processing. However, despite the current advances in GPU languages and tools, taking advantage of their parallel architecture is still far more complex than programming standard multi-core CPUs. Performance profiling and analysis of known applications provides a useful insight of the hardware architecture and memory hierarchy. Afterwards, this analysis can be used to identify potential bottlenecks and tune other software so it can make a more efficient usage of the available resources. In this work we implement a small signal processing library which will be used to characterize the performance of most recent NVIDIA GPU architectures. The methodology used in our signal processing library is based on a series of building blocks that enable us to easily design several well-known algorithms with little effort. The library was built paying special attention to flexibility and adaptability. In this work we also show how a generic approach can be used to easily design these GPU algorithms while obtaining competitive performance, which results specially interesting from the productivity standpoint., This research has been economically supported by Xunta de Galicia under project CN2012/211, cofunded by FEDER funds of the European Union under the grant TIN2010–16735., Xunta de Galicia; CN2012/211
RUC. Repositorio da Universidade da Coruña
Publicación Artículo científico (JournalArticle)

Parallel Monte Carlo radiosity using scene partitioning

  • Sanjurjo Amado, José Rodrigo
  • Amor, Margarita
  • Bóo, Montserrat
  • Doallo, Ramón
This is an accepted version of: Sanjurjo JR, Amor M, Bóo M, Doallo R. Parallel Monte Carlo radiosity using scene partitioning. The International Journal of High Performance Computing Applications. 2013;27(3):318-334. Copyright © 2013, Sage Publications. DOI: 10.1177/1094342012457191, [Abstract]: Global illumination provides realistic image synthesis but its high computational requirements limit its use in practice. In this paper we present a parallel method of the Monte Carlo radiosity method. Our proposal is based on the utilization of a convex partition to divide the whole scene into a set of disjoint sub-scenes which are allocated among processors of a distributed memory system. We have used two partitioning strategies: uniform and non-uniform. The convex partition employed permits the exploitation of data locality, and the optimization of the ray shooting procedure by minimizing the number of objects to be tested in the intersection calculation. We present three different techniques to increase the performance of parallel implementation and to solve the challenges that the distribution of the data among processors implies: minimization of communications, load balancing, and a distributed test for determining the end of each iteration. The obtained results are good in terms of quality and execution times, increasing the flexibility of previous solutions., This work was partially supported by the Ministry of Education and Science of Spain under the contract MEC TIN 2010-16735 and also supported by the Xunta de Galicia under the contracts INCITE08PXIB105161PR, 08TIC001206PR., Xunta de Galicia; INCITE08PXIB105161PR
RUC. Repositorio da Universidade da Coruña
Publicación Artículo científico antes de ser publicado, versión del autor (Preprint)

Exploiting Heterogeneous Parallelism With the Heterogeneous Programming Library

  • Viñas Buceta, Moisés
  • Bozkus, Zeki
  • Fraguela, Basilio B.
[Abstract] While recognition of the advantages of heterogeneous computing is steadily growing, the issues of programmability and portability hinder its exploitation. The introduction of the OpenCL standard was a major step forward in that it provides code portability, but its interface is even more complex than that of other approaches. In this paper, we present the Heterogeneous Programming Library (HPL), which permits the development of heterogeneous applications addressing both portability and programmability while not sacrificing high performance. This is achieved by means of an embedded language and data types provided by the library with which generic computations to be run in heterogeneous devices can be expressed. A comparison in terms of programmability and performance with OpenCL shows that both approaches offer very similar performance, while outlining the programmability advantages of HPL., This work was funded by the Xunta de Galicia under the project “Consolidación e Estructuración de Unidades de Investigación Competitivas” 2010/06 and the MICINN, cofunded by FEDER funds, under grant TIN2010-16735. Zeki Bozkus is funded by the Scientific and Technological Research Council of Turkey (TUBITAK; 112E191), Scientific and Technological Research Council of Turkey (TUBITAK); 112E191
RUC. Repositorio da Universidade da Coruña
Publicación Artículo científico (JournalArticle)

Numerical Simulation of Pollutant Transport in a Shallow-Water System on the Cell Heterogeneous Processor

  • González, Carlos H.
  • Fraguela, Basilio B.
  • Andrade, Diego
  • García Rodríguez, José Antonio
  • Castro, M.J.
[Abstract] This paper presents an implementation, optimized for the Cell processor, of a finite volume numerical scheme for 2D shallow-water systems with pollutant transport. A description of the special architecture and programming required by the Cell processor motivates the methodology to develop optimized implementations for this platform. This process involves parallelization, data structure reorganization, explicit transfers of data and computation vectorization. Our implementation, tested using a realistic problem, achieves very good speedups with respect to the sequential execution on a standard CPU., This work was partially supported by the Science and Innovation Ministry of Spain (Projects TIN2010-16735, MTM2010-21135-C02-01 and MTM2009-11923), Xunta de Galicia CN2012/211 (partially supported by FEDER funds), and the FPU program of the Spanish Government (ref AP2009-4752), Xunta de Galicia; CN 2012/211
RUC. Repositorio da Universidade da Coruña
Publicación Artículo científico (JournalArticle)

Efficient Adaptive and Dynamic Mesh Refinement Based on a Non-recursive Strategy

  • Bóo, Montserrat
  • Amor, Margarita
  • Concheiro, Raquel
  • Doggett, Michael
This is a pre-copyedited, author-produced version of an article accepted for publication in The Computer Journal following peer review. The version of record M. Bóo, M. Amor, R. Concheiro, M. Doggett, Efficient Adaptive and Dynamic Mesh Refinement Based on a Non-recursive Strategy, The Computer Journal, Volume 56, Issue 7, July 2013, Pages 843–851, is available online at: https://academic.oup.com/comjnl/article-abstract/56/7/843/397484?redirectedFrom=fulltext [https://doi.org/10.1093/comjnl/bxs046]., [Abstract]: In this paper, we present a meshing scheme for the implementation of an adaptive tessellation of triangular meshes on the graphics processing unit (GPU). Tessellation is performed according to a local test to generate primitives dynamically. The refinement procedure does not require the pre-computation of any refinement pattern. The resulting adaptive procedure is efficient and simple, and generates the tessellation pattern of each triangle dynamically. The non-recursive strategy simplifies mesh reconstruction, avoids irregular memory access patterns and uses a simple control flow to make it a good candidate to guide the evolution of tessellation algorithms on future graphics cards. To test our proposal, we have used the geometry shader of current GPUs to generate primitives dynamically., This work was supported by the Xunta de Galicia under the Program for Consolidation of Competitive Research Groups ref. 2010/06 and by the Ministry of Science and Innovation of Spain, cofunded by the FEDER funds of the European Union, under contract TIN2010-16735., Xunta de Galicia; 2010/06
RUC. Repositorio da Universidade da Coruña
Publicación Artículo científico (JournalArticle)

Extended Hybrid Meshing Algorithm for Multiresolution Terrain Models

  • Paredes, E.G
  • Bóo, Montserrat
  • Amor, Margarita
  • Bruguera, Javier D.
  • Döllner, Jürgen
This is an Accepted Manuscript of an article published by Taylor & Francis in International Journal of Geographical Information Science on 03 Nov 2011, available at: https://doi.org/10.1080/13658816.2011.615317, [Abstract]: Hybrid terrains are a convenient approach for the representation of digital terrain models, integrating heterogeneous data from different sources. In this article, we present a general, efficient scheme for achieving interactive level-of-detail rendering of hybrid terrain models, without the need for a costly preprocessing or resampling of the original data. The presented method works with hybrid digital terrains combining regular grid data and local high-resolution triangulated irregular networks. Since grid and triangulated irregular network data may belong to different datasets, a straightforward combination of both geometries would lead to meshes with holes and overlapping triangles. Our method generates a single multiresolution model integrating the different parts in a coherent way, by performing an adaptive tessellation of the region between their boundaries. Hence, our solution is one of the few existing approaches for integrating different multiresolution algorithms within the same terrain model, achieving a simple interactive rendering of complex hybrid terrains., This work was supported in part by the Xunta de Galicia under projects 08TIC001206PR and 08SIN011291PR and the Program for Consolidation of Competitive Research Groups ref. 2010/06 and 2010/28, and by the Ministry of Science and Innovation of Spain, cofunded by the FEDER funds of the European Union, under contracts TIN2010-16735 and TIN2010-17541. M. Bóo, M. Amor, and J.D. Bruguera are members of the European Network of Excellence on High Performance and Embedded Architecture and Compilation (HiPEAC)., Xunta de Galicia; 08TIC001206PR, Xunta de Galicia; 08SIN011291PR
RUC. Repositorio da Universidade da Coruña
Publicación Artículo científico (JournalArticle)

Influence of memory access patterns to small-scale FFT performance

  • Lobeiras Blanco, Jacobo
  • Amor, Margarita
  • Doallo, Ramón
This version of the article has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: http://dx.doi.org/10.1007/s11227-012-0807-5, [Abstract]: Modern GPUs (Graphics Processing Units) offer very high computing power at relative low cost. To take advantage of their computing resources and develop efficient implementations is essential to have certain knowledge about the architecture and memory hierarchy. In this paper, we use the FFT (Fast Fourier Transform) as a benchmark tool to analyze different aspects of GPU architectures, like the influence of the memory access pattern or the impact of the register pressure. The FFT is a good tool for performance analysis because it is used in many digital signal processing applications and has a good balance between computational cost and memory bandwidth requirements., This work was supported by the Xunta de Galicia under projects 08TIC001206PR and INCITE08PXIB105161PR, the Ministry of Science and Innovation, cofunded by the FEDER funds of the European Union under the grant TIN2010-16735, and the Consolidation of Competitive Research Groups ref. 2010/06., Xunta de Galicia; 08TIC001206PR, Xunta de Galicia; INCITE08PXIB105161PR