Extending OpenMP to Survive the Heterogeneous Multi-Core Era

被引:32
|
作者
Ayguade, Eduard [1 ,2 ]
Badia, Rosa M. [1 ,3 ]
Bellens, Pieter [1 ]
Cabrera, Daniel [1 ]
Duran, Alejandro [1 ]
Ferrer, Roger [1 ]
Gonzalez, Marc [1 ,2 ]
Igual, Francisco [4 ]
Jimenez-Gonzalez, Daniel [1 ,2 ]
Labarta, Jesus [1 ,2 ]
Martinell, Luis [1 ]
Martorell, Xavier [1 ,2 ]
Mayo, Rafael [4 ]
Perez, Josep M. [1 ]
Planas, Judit [1 ]
Quintana-Orti, Enrique S. [4 ]
机构
[1] Barcelona Supercomp Ctr, Ctr Nacl Supercomputac BSC CNS, Barcelona 08034, Spain
[2] Univ Politecn Cataluna, Depto Arquitectura Comp, ES-08034 Barcelona, Spain
[3] Spanish Natl Res Council, CSIC, Artificial Intelligence Res Inst, Madrid, Spain
[4] UJI, Depto Ingn & Ciencia Comp, Castellon de La Plana 12071, Spain
关键词
Parallel computing; Programming models; Runtime systems; Task-level parallelism; Multi-core processors; Hardware accelerators; Heterogeneous computing; DESIGN;
D O I
10.1007/s10766-010-0135-4
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
This paper advances the state-of-the-art in programming models for exploiting task-level parallelism on heterogeneous many-core systems, presenting a number of extensions to the OpenMP language inspired in the StarSs programming model. The proposed extensions allow the programmer to write portable code easily for a number of different platforms, relieving him/her from developing the specific code to off-load tasks to the accelerators and the synchronization of tasks. Our results obtained from the StarSs instantiations for SMPs, the Cell, and GPUs report reasonable parallel performance. However, the real impact of our approach in is the productivity gains it yields for the programmer.
引用
收藏
页码:440 / 459
页数:20
相关论文
共 50 条
  • [1] Extending OpenMP to Survive the Heterogeneous Multi-Core Era
    Eduard Ayguadé
    Rosa M. Badia
    Pieter Bellens
    Daniel Cabrera
    Alejandro Duran
    Roger Ferrer
    Marc Gonzàlez
    Francisco Igual
    Daniel Jiménez-González
    Jesús Labarta
    Luis Martinell
    Xavier Martorell
    Rafael Mayo
    Josep M. Pérez
    Judit Planas
    Enrique S. Quintana-Ortí
    International Journal of Parallel Programming, 2010, 38 : 440 - 459
  • [2] The Future of OpenMP in the Multi-Core Era
    Chapman, Barbara
    Mey, Dieter An
    PARALLEL COMPUTING: ARCHITECTURES, ALGORITHMS AND APPLICATIONS, 2008, 15 : 571 - +
  • [3] Extending Synchronization Constructs in OpenMP to Exploit Pipeline Parallelism on Heterogeneous Multi-core
    Li, Shigang
    Yao, Shucai
    He, Haohu
    Sun, Lili
    Chen, Yi
    Peng, Yunfeng
    ALGORITHMS AND ARCHITECTURES FOR PARALLEL PROCESSING, PT II, 2011, 7017 : 54 - 63
  • [4] OpenMDSP: Extending OpenMP to Program Multi-Core DSPs
    He, Jiang-Zhou
    Chen, Wen-Guang
    Chen, Guang-Ri
    Zheng, Wei-Min
    Tang, Zhi-Zhong
    Ye, Han-Dong
    JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY, 2014, 29 (02) : 316 - 331
  • [5] OpenMDSP:Extending OpenMP to Program Multi-Core DSPs
    何江舟
    陈文光
    陈光日
    郑纬民
    汤志忠
    叶寒栋
    Journal of Computer Science & Technology, 2014, 29 (02) : 316 - 331
  • [6] OpenMDSP: Extending OpenMP to Program Multi-Core DSPs
    Jiang-Zhou He
    Wen-Guang Chen
    Guang-Ri Chen
    Wei-Min Zheng
    Zhi-Zhong Tang
    Han-Dong Ye
    Journal of Computer Science and Technology, 2014, 29 : 316 - 331
  • [7] Implementation and optimization of OpenMP task parallelism on heterogeneous multi-core architecture
    Li, Shigang
    Hu, Changjun
    Journal of Computational Information Systems, 2013, 9 (12): : 4981 - 4988
  • [8] Loading OpenMP to Cell: An effective compiler framework for heterogeneous multi-core chip
    Wei, Haitao
    Yu, Junqing
    PRACTICAL PROGRAMMING MODEL FOR THE MULTI-CORE ERA, PROCEEDINGS, 2008, 4935 : 129 - 133
  • [9] Heterogeneous multi-core architectures
    Mitra, Tulika
    IPSJ Transactions on System LSI Design Methodology, 2015, 8 : 51 - 62
  • [10] A Performance model for OpenMP programs on multi-core machines
    Nasri, Wahid
    Fathallah, Karim
    2013 INTERNATIONAL CONFERENCE ON COMPUTER APPLICATIONS TECHNOLOGY (ICCAT), 2013,