A parallel implementation study of Gaussian Elimination using OpenMP and CUDA

  • Vitor Milanez UFSCar
  • Ivan Navas UFSCar
  • João Migliatti UFSCar
  • Matheus Miata UFSCar
  • Hélio Guardia UFSCar
DOI: https://doi.org/10.5753/eradsp.2024.239916

Abstract


With the increase in the degree of parallelism in computer systems, interest in parallel implementations to reduce execution times for the most varied problems also increases. Knowing this, in this article we present two parallel versions for the Gaussian Elimination problem, one in a multi-core environment and the other using a GPU. The results obtained show parallelism is advantageous in this case, with greater gains obtained with GPU.

References

S.F.MCGINN; R.E.SHAW. Parallel Gaussian Elimination Using OpenMP and MPI. 16Th Annual International Symposium On High Performance Computing Systems And Applications. Saint John, p. 1-5. jun. 2002.

FACCI, Henrique B.; GONÇALVES, Ronaldo A. de L. EXECUÇÃO DE SISTEMAS. DE EQUAÇÕES LINEARES EM PROCESSADORES MULTI-CORE. IV Epac: Encontro Paranaense de Computação. Maringá, p. 1-9. maio 2011.

GRCAR, Joseph F.. Mathematicians of Gaussian Elimination. Notices Of The Ams. Providence, Ri, p. 782-792. jul. 2011.

IEEE (org.). 5th Marathon of Parallel Programming. SBAC-PAD’2010. Petrópolis, Rio de Janeiro, p. 5-6. 28 out. 2010.
Published
2024-05-16
How to Cite

Select a Format
MILANEZ, Vitor; NAVAS, Ivan; MIGLIATTI, João; MIATA, Matheus; GUARDIA, Hélio. A parallel implementation study of Gaussian Elimination using OpenMP and CUDA. In: REGIONAL SCHOOL OF HIGH PERFORMANCE COMPUTING FROM SÃO PAULO (ERAD-SP), 15. , 2024, Rio Claro/SP. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2024 . p. 29-32. DOI: https://doi.org/10.5753/eradsp.2024.239916.