Analysis of Programs for GPGPU Architectures
Lakshmi M. Gadhikar, Y. S. Rao · 2018 2nd International Conference on Trends in Electronics and Informatics (ICOEI) · 2018
Over the past decade, General Purpose Graphics Procesing Units i.e. GPGPUs emerged as a platform for general purpose and scientific computing applications. Languages used for programming GPGPUs are CUDA, OpenCL, DirectCompute etc. As GPUs are widely being used in scientific computing, optimization of programs has become a major concern. Producing accurate results in reduced time is increasingly becoming an important factor in the usability of GPU applications in concerned domains. Program optimization requires analysis of program to find the program features or parameters that impact its execution and storage efficiency. In this work, we analyze GPGPU programs written in CUDA to extract important features in them which will be useful for further optimization.