Powiadomienia systemowe
- Sesja wygasła!
- Sesja wygasła!
- Sesja wygasła!
- Sesja wygasła!
Identyfikatory
Warianty tytułu
Użycie GPU w celu zwiększenia wydajności obliczania recurrence plot
Języki publikacji
Abstrakty
Simulation and analysis of sophisticated systems require much computations. Moore’s law, although still allows for increasing number of transistors on the die, does not lead to increase of performance of single chip — instead it leads to increased parallelism of entire system. This allows for improving performance of those algorithms that can be parallelised; recurrence plot is one of such algorithms. Graphical Processing Units (GPU) show the largest increase of parallel computations capabilities. At the same time they do not behave as traditional CPUs and require different style of programming to fully utilise their capabilities. Article shows techniques that can be used to increase performance of computing of recurrence plot on GPGPU.
Analiza skomplikowanych systemów wymaga przeprowadzenia wielu obliczeń. Prawo Moore’a, choć wciąż˙ pozostaje w mocy, nie pozwala na zwiększanie wydajności pojedynczego procesora, ale pomaga w tworzeniu wydajnych równoległych systemów. Pozwala to na zwiększanie wydajności dla algorytmów które można zrównoleglić; recurrence plot należy do takich algorytmów. Procesory graficzne (GPU) oferują największą ilość równoległych jednostek obliczeniowych, jednocześnie jednak ich wydajne wykorzystanie wymaga innego podejścia programistycznego. Artykuł opisuje w jaki sposób wykorzystano technologię CUDA do przyśpieszania obliczania recurrence plot.
Rocznik
Tom
Strony
77--94
Opis fizyczny
Bibliogr. 15 poz., tab., wykr.
Twórcy
autor
- Politechnika Białostocka, Wydział Informatyki, Białystok
Bibliografia
- [1] Peter Bakkum and Kevin Skadron. Accelerating sql database operations on a gpu with cuda. In Proceedings of the 3rd Workshop on General-Purpose Computation on Graphics Processing Units, GPGPU ’10, pages 94–103, New York, NY, USA, 2010. ACM.
- [2] ATI Corporation. ATI Stream Computing OpenCL Programming Guide, 08 2010. http://developer.amd.com/gpu/ATIStreamSDK/assets/ATI_Stream_ SDK_OpenCL_Programming_Guide.pdf Accessed at 2010-10-04.
- [3] NVIDIA Corporation. NVIDIA CUDA C Best Practices Guide Version 3.2, 08 2010. http://developer.download.nvidia.com/compute/cuda/3_2/ toolkit/docs/CUDA_C_Best_Practices_Guide.pdf Accessed at 2010-10-04.
- [4] NVIDIA Corporation. NVIDIA CUDA C Programming Guide Version 3.2, 09 2010. http://developer.download.nvidia.com/compute/cuda/3_2/ toolkit/docs/CUDA_C_Programming_Guide.pdf Accessed at 2010-10-04.
- [5] NVIDIA Corporation. NVIDIA CUDA Reference Manual Version 3.2, 08 2010. http://developer.download.nvidia.com/compute/cuda/3_2/toolkit/docs/CUDA_ Toolkit_Reference_Manual.pdf Accessed at 2010-10-04.
- [6] NVIDIA Corporation. NVIDIA OpenCL Programming Guide for the CUDA Architecture Version 3.2, 08 2010. http://developer.download.nvidia.com/ compute/cuda/3_2/toolkit/docs/OpenCL_Programming_Guide.pdf Accessed at 2010-10-04.
- [7] NVIDIA Corporation. Tuning CUDA Applications for Fermi Version 1.3, 08 2010. http://developer.download.nvidia.com/compute/cuda/3_2/toolkit/ docs/Fermi_Tuning_Guide.pdf Accessed at 2010-10-04.
- [8] Kayvon Fatahalian and Mike Houston. A closer look at gpus. Communications of ACM, 51(10):50–57, 2008.
- [9] Erich Gamma, Richard Helm, Ralph Johnson, and John Vlissides. Design Patterns. Elements of Reusable Object–Oriented Software. Addison–Wesley, 1995.
- [10] Liang Gu, Xiaoming Li, and Jakob Siegel. An empirically tuned 2d and 3d fft library on cuda gpu. In Proceedings of the 24th ACM International Conference on Supercomputing, ICS ’10, pages 305–314, New York, NY, USA, 2010. ACM.
- [11] Andreas Klöckner, Nicolas Pinto, Yunsup Lee, Bryan Catanzaro, Paul Ivanov, and Ahmed Fasih. Pycuda: Gpu run-time code generation for high-performance computing.
- [12] Allen D. Malony, Scott Biersdorff,Wyatt Spear, and Shangkar Mayanglambam. An experimental approach to performance measurement of heterogeneous parallel applications using cuda. In Proceedings of the 24th ACM International Conference on Supercomputing, ICS ’10, pages 127–136, New York, NY, USA, 2010. ACM.
- [13] Akira Nukada and Satoshi Matsuoka. Auto-tuning 3-d fft library for cuda gpus. In Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC ’09, pages 30:1–30:10, New York, NY, USA, 2009. ACM.
- [14] Tomasz Rybak and Romuald Mosdorf. Computer users activity analysis using recurrence plot. In International Conference on Biometrics and Kansei Engineering, Cieszyn, Poland, 2009. AGH.
- [15] Tomasz Rybak and Romuald Mosdorf. User activity detection in computer systems by means of recurrence plot analysis. Zeszyty Naukowe Politechniki Białostockiej. Informatyka Zeszyt 5, pages 67–86, 2010.
Typ dokumentu
Bibliografia
Identyfikator YADDA
bwmeta1.element.baztech-article-BPB1-0047-0019