Energy Efficient Computation Method for CPU-GPU System on Chip

C. Sai Punitha
2018 International Journal for Research in Applied Science and Engineering Technology  
The increasing trends in multi-core chips allows higher performance at lower energy and the communication between the cores is a limiting factor which can be improved by the parallel computation such as thread level parallelism. The improvement in performance gained by the use of a multi-core processor depends very much on the software algorithms used and their implementation. In particular, possible gains are limited by the fraction of the software that can run in parallel simultaneously on
more » ... tiple cores. One such algorithm called Double Modulus Number Theoretic Transform finds good for large amount of data computational methods especially with zero round off errors which provides high efficiency in communication between heterogeneous multicores The parallelization algorithm Double Modulus Number Theoretic Transform is computed and developed with software development platform CUDA especially design for its Nvidia Graphic cards which provides higher bandwidth and speed with less computational complexity.
doi:10.22214/ijraset.2018.5199 fatcat:g3wqyatcdffa3n743oazsds7wi