In the past decade, high performance compute capabilities exhibited by heterogeneous GPGPU platforms have led to popularity of data parallel programming languages such as CUDA and OpenCL. Developing solutions using involve a steep learning curve due complexity underlying devices their impact on performance. This has emergence several High Performance Computing frameworks which provide high-leve...

The aim of this study will be to design Parallel solver (PS) for oscillatory stiff systems ordinary differential equations (ODEs). PS constructed via a type specially transformed exponentially fitted multinomial approximant in accordance with the behaviour solution. method interpolation and collocation utilized. principal local truncation errors used derive suitable step size decide error toler...