Performance modeling and analysis for many-core processors part 1(GPU)

• 主要作者: Seet, Ivan Wei Ren.
• 其他作者: School of Computer Engineering
• 格式: Final Year Project
• 語言: English
• 出版: 2013
• 主題: DRNTU::Engineering::Computer science and engineering
• 在線閱讀: http://hdl.handle.net/10356/52793

Data analyze has become very important with growth of information today. There is a need of real-time information to be process and analyze fast. This led to the research analyze of the current many-core processor which can support fast processing of large amount of data. Graphic Processor Unit (GPU) has emerged as one of the many-core processor in the current market. With the high parallel multithreading, the many-core processor with tremendous computational horsepower and high memory bandwidth it provide a good platform to process large amount of data when the program design for the process leverage the architecture of the GPU. Sorting is chosen as one of the process to study and research on the performance of the GPU architecture. Therefore in this report it will go through the different type of sort technique available and it parallel algorithm. The NVIDIA GPU Fermi and Kepler architecture have also been chosen as the platform where these sort algorithms will model on and analyze result will perform to evaluate it performance.