Implementation of Cubic Spline Interpolation on Parallel Skeleton Using Pipeline Model on CPU-GPU Cluster
Authors:
- Prasant Kumar MohantyPMPrasant Kumar Mohanty
- Motahar RezaMRMotahar Reza
- Piyush KumarPKPiyush Kumar
- Praveen KumarPKPraveen Kumar
- 1 February 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 747-751
Abstract
The cubic spline interpolation is frequently used for analysis the data set in various aspects of engineering and science problem. For a large set of data points defined with very large range, it is very difficult to interpolate by using traditional sequential algorithm. In this paper, we proposed a more systematic approach which has a parallel component known as skeleton which is implemented in various parallel paradigms like OpenMP, MPI, and CUDA etc. It is interesting that the skeleton approach is used with pipelining technique that gives better result as compared to the previous studies. This approach is applied to compute the cubic spline interpolating polynomial based on a large data set. The experimental result using the parallel skeleton technique on multi-core CPU and GPU shows better performance with respect to other parallel methods.Keywords
This publication has 7 references indexed in Scilit:
- Parallel direct methods for solving the system of linear equations with pipelining on a multicore using OpenMPJournal of Computational and Applied Mathematics, 2011
- Fast tridiagonal solvers on the GPU2010
- A cost-optimal parallel implementation of a tridiagonal system solver using skeletonsFuture Generation Computer Systems, 2005
- Cyclic reduction on distributed shared memory machines2001
- Unified architecture for divide and conquer based tridiagonal system solversIEEE Transactions on Computers, 1994
- A Parallel Method for Tridiagonal EquationsACM Transactions on Mathematical Software, 1981
- An Efficient Parallel Algorithm for the Solution of a Tridiagonal Linear System of EquationsJournal of the ACM, 1973