Performance Optimization of Transformation Technique - DCT using NVIDIA CUDA

Abstract

GPU- graphics processing unit play a major role in many computational environments, most notably those regarding real-time graphics applications, such as image processing. CUDA programmed GPUs are rapidly becoming a major choice in high performance computing and there are a growing number of applications which are being ported to the CUDA platform. However much less research has been carried out to evaluate the optimized performance, when CUDA is integrated with other parallel programming paradigms. The goal of this project is to explore the potential performance improvements that could be gained through the use of GPU processing techniques within the NVIDIA CUDA architecture. In this paper we have implemented DCT Compression algorithm in CUDA. Most of CPU-based implementations of DCT are firmly adjusted for operating using fixed point arithmetic but still appear to be rather costly as soon as blocks are processed in the sequential order by the single ALU. Performing DCT computation on GPU using NVIDIA CUDA technology gives significant performance boost even compared to a modern CPU.