期刊文献+

异构多核图形处理器存储系统设计与实现 被引量:4

Design and implementation of memory system in heterogeneous multi-core GPU
下载PDF
导出
摘要 提出了异构多核图形处理器(HMGPU)存储管理系统的硬件实现方法,采用固定分区与分页式分区两种方式分别对大片连续数据与小片非连续数据进行管理,使用Verilog语言进行硬件设计和仿真,并在FPGA开发板上进行了验证。实验结果表明,该系统为HMGPU提供了2 021.2 MB/s的有效存储带宽。 A hardware method for heterogeneous muhi-core graphic processor unit(HMGPU) memory management system is pro-posed in this paper. Fixed-partition is used to manage big-block and continuous data ,and paging management methods is used to manage small-block and non-continuous data.The memory management system is designed and simulated in Verilog, and it is verified to be valid in FPGA. Experimental results show that the system can provide effectively 2 021.2 MB/s memory bandwidth for HMGPU.
出处 《电子技术应用》 北大核心 2013年第5期38-40,43,共4页 Application of Electronic Technique
基金 国家自然科学基金重点项目(61136002)
关键词 异构多核图形处理器 实时并行 硬件实现方法 存储管理系统 heterogeneous multi-core graphic processor real-time and parallel hardware method memory management system
  • 相关文献

参考文献6

  • 1COPE B,PETER Y K,LUK W.Using reconfigurable logic to optimise GPU memory accesses[C].Design,Automation and Test in Europe, DATE'08.Munich, 2008:44-49.
  • 2马安国,成玉,唐遇星,邢座程.GPU异构系统中的存储层次和负载均衡策略研究[J].国防科技大学学报,2009,31(5):38-43. 被引量:11
  • 3HENRY W, MISELMYRTO P, MARYAM S A,et al.Demystifying GPU microarchitecture through microbenchmarking[C]. Performance Analysis of Systems & Software, IEEE International Symposium on Computing & Processing(Hardware/ Software), White Plains, NY, March 2010 : 235-246.
  • 4NVIDIA Corporation.NVIDIA CUDA compute unified device architecture reference manual[EB/OL].(2008-06-xx)[2013- 01 - 15 ]. http ://www. cs. ucla. edu/- palsberg/course/cs239/ papers / CudaReferenceManual 2.0. pdf. 1 - 247.
  • 5r MCCORMACK J, MCNAMARA R, GIANOS C, et al.Implementing neon:a 256-bit graphics accelerator[J].Micro IEEE, 1999,19(2) : 58-69.
  • 6MCCORMACK J, MCNAMARA R, GIANOS C, et al.Neon : A single-chip 3D workstation graphics accelerator,research report 98/1[C].HWWS'98 Proceedings of the ACM SIG- GRAPH/EUROGRAPHICS Workshop on Graphics Hardware ACM, New York, NY, USA, 1998 : 123-132.

二级参考文献10

  • 1Hartley T D R,Catalyurek U,Ruiz A,et al.Biomedical Image Analysis on A Cooperative Cluster of Gpus and Multicores[].Proceedings of thendAnnual International Conference on Supercomputing.2008
  • 2Bond A.Havok F X:GPU-accelerated Physics for PC Games[].Proceedings of Game Developers Conference.2006
  • 3Hagen T R,Lie K A,Natvig J R.Solving the Euler Equations on Graphics Processing Units[].Proceedings of the thInternational Conference onComputational Science.2006
  • 4Elsen E,Houston M,Vishal V,et al.BN-body Simulation on GPUs[].Proc ACM/IEEE Confon Supercomputing.2006
  • 5Stone S S,Haldar J P,Tsao S C,et al.Accelerating Advanced MRI Reconstructions on GPUs[].ACM Computing Frontier Conference.2008
  • 6.OpenVIDIA:GPU-accelerated Computer Vision Library[EB][].openvidiasourceforgenet.2006
  • 7Volkov V,Demmel J W.Benchmarking GPUs to Tune Dense Linear Algebra[].SC’:Proceedings of the ACM/IEEE Conference on Su-per-computing.2008
  • 8Fatica M.Accelerating Linpack with CUDA on Heterogenous Clusters[].GPGPU’.2009
  • 9Halfhill T R.Parallel Processing with CUDA[].Microprocessor Report.2008
  • 10Gutowitz H.A Tutorial Introduction to Swarm[]..1993

共引文献10

同被引文献16

引证文献4

二级引证文献11

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部