Implementation of Airy function using Graphics Processing Unit (GPU)

Abstract

Special mathematical functions are an integral part of Fractional Calculus, one of them is the Airy function. But it’s a gruelling task for the processor as well as system that is constructed around the function when it comes to evaluating the special mathematical functions on an ordinary Central Processing Unit (CPU). The Parallel processing capabilities of a Graphics processing Unit (GPU) hence is used. In this paper GPU is used to get a speedup in time required, with respect to CPU time for evaluating the Airy function on its real domain. The objective of this paper is to provide a platform for computing the special functions which will accelerate the time required for obtaining the result and thus comparing the performance of numerical solution of Airy function using CPU and GPU.

Analysis of GPU Computation of Parabolic, Bessel, Wright and Riemann Zeta Functions

“…Run the code using Run & Time Function. 4. Analyze and Convert the section that takes more time to execute on GPU.…”

“…Computation of such functions has proven to be a resource and time intensive task. To tackle this problem, coprocessors such as GPU and parallel algorithms have been used by researchers in the past [4]. The general purpose use of GPUs for computing is commonly known as GPGPU (General Purpose Computing on Graphics Processing Unit) [3].…”