FINITE ELEMENT MATRIX GENERATION ON A GPU

Citations Over TimeTop 10% of 2012 papers

Abstract

This paper presents an efficient technique for fast generation of sparse systems of linear equations arising in computational electromagnetics in a finite element method using higher order elements. The proposed approach employs a graphics processing unit (GPU) for both numerical integration and matrix assembly. The performance results obtained on a test platform consisting of a Fermi GPU (1x Tesla C2075) and a CPU (2x twelve-core Opterons), indicate that the GPU implementation of the matrix generation allows one to achieve speedups by a factor of 81 and 19 over the optimized singleand multi-threaded CPU-only implementations, respectively.

Related Papers

• → Interprocedural parallelization analysis in SUIF(2005)66 cited
• → Parallelization of a dynamic unstructured algorithm using three leading programming paradigms(2000)27 cited
• → Automatic parallelization for non-cache coherent multiprocessors(1997)4 cited
• Multi-core Parallel Computational Model Based on Horizontal Locality(2012)
• → Speculative parallelization on GPGPUs(2012)