HIGH PERFORMANCE COMPUTING SOLUTIONS WITH CUDA TECHNOLOGY

Kaspars Vogulis, Sergejs Kodors

Abstract


Proposed research is completed to view the data computing capabilities with CUDA technology and video card options. The goal of work is to compare task execution speed using the CPU and CUDA technology. Research study concluded that the use of CUDA technology in certain tasks can improve execution time and save system resources. CUDA technology uses the graphic processor (GPU) parallel architecture that allows a large number of tasks to be solved simultaneously and independently of each other.

Keywords


CUDA technology; computing operations; data operations; graphics processor

Full Text:

PDF

References


Nickolls J., Kirk D. Graphics and Computing GPUs. Computer organization and design Amerikas Savienotās Valstis: Elsevier INC.

Patterson D.A.,. Hennesy J.L,Computer organization and design Amerikas Savienotās Valstis: Elsevier INC. 2014.gads

Sanders J., Kandrot E. CUDA by Example. Amerikas Savienotās valstis. Nvidia . 2011. gads

Nickolls J., Dally W.J. THE GPU COMPUTING ERA. 2010.gads. Amerikas Savienotās valstis. IEEE Computer Society. Nvidia. 2010.gads

Garland M, Kudlur M., Zheng Y. Designing a Unified Programming Model for Heterogeneous Machines. Amerikas Savienotās valstis. Nvidia. 2010.gads

Cook S. CUDA Programming. A Developer’s Guide to Parallel computing with GPUs. Amerikas Savienotās valstis. Elsevier INC. 2013.gads

Computational Fluid Dynamics. Sk. internetā (21.05.2016) http://www.nvidia.com/object/national_center_for_atmospheric_research.html

NATO CMRE REVOLUTIONIZES REAL-TIME UNDERSEA MINE DETECTION. Sk internetā (19.05.2016) http://www.nvidia.com/content/tesla/pdf/nato-case-study.pdf

Farber R. CUDA Application Design and Development.Amerikas Savienotās valstis. Elsevier INC. 2011.gads




DOI: http://dx.doi.org/10.17770/het2017.21.3566

Refbacks

  • There are currently no refbacks.