DECK-GM: Uma Implementação do Ambiente DECK para Myrinet
Resumo
O uso da tecnologia Myrinet como um padrão em agregados torna necessário o desenvolvimento de bibliotecas de comunicação que explorem suas características. Esse artigo apresenta a modelagem e a implementação do ambiente DECK para Myrinet, utilizando a API do sistema GM. Serão apresentadas as características do DECK, do GM e será detalhada a implementação mostrando os mecanismos utilizados para que se pudesse alcançar o zero-copy.
Referências
M. Barreto, R. Ávila, R. Cassali, A. Carissimi, and P. Navaux. Implementation of the DECK environment with BIP. In Proc. of the 1st Myrinet User Group Conference, pages 82-88. Lyon, France, 2000. Lyon. INRIA Rocquencourt.
M. Barreto, R. Ávila, F. de Oliveira, R. Cassali, and P. Navaux. DECK: an enviroment for parallel programming on clusters of multiprocessors. In Proc. of the 12th Symposium on Computer Architecture and High-Performance Computing, pages 121 - 329. São Pedro, SP, 2000. São Carlos. UFSCAR.
M. Barreto, R. Ávila, and P. Navaux. The MultiCluster model to the integrated use of multiple workstation clusters. In J. Rolim et al., editors. Proc. of the 3rd Workshop on Performance Computer based Networks of Workstations, volume 1800 of Lecture Notes in Computer Science, pages 71-80. Cancun, 2000. Berlin. Springer-Verlag.
M. E. Barreto, P. O. A. Navaux, and M. P. Rivière. DECK: a new model for a distributed executive kernel integrating communication and multithreading for support of distributed object oriented application with fault tolerance support. In Trabajos Seleccionados del 4o. Congreso Argentino de Ciencias de la Computación, volume 2, pages 623-637. Neuquén, Argentina, 1998. Neuquén. Universidad Nacional del Comahue.
N. Boden et al., Myrinet: a gigabit-per-second local-area network. IEEE Micro. 15(1):29-36. Feb. 1995.
F. A. D. de Oliveira. Uma biblioteca para programação paralela por troca de mensagens de clusters baseados na tecnologia SCI. Master's thesis. Instituto de Informática. Universidade Federal do Rio Grande do Sul. Porto Alegre. 2001.
F. A. D. de Oliveira, R. B. A vila, M. E. Barreto, and P. O. A. Navaux. Low-latency and zero-copy message passing protocols for SCI-based clusters. In A. C. M. A. Melo, M. A. R. Dantas, and J. Panetta, editors, Proc. of the 13th Symposium on Computer Architecture and High-Performance Computing, pages 148-155. Pirenópolis, GO, 2001. Brasília. Departamento de Ciência da Computação da UNB.
F. A. D. de Oliveira, R. B. Ávila, M. E. Barreto, P. O. A. Navaux, and C. De Rose. DECK-SCI: High-performance communication and multithreading for SCI clusters. In D. S. Katz, T. Sterling, M. Baker, L. Bergman, M. Paprzyeki, and R. Buyya, editors. Proc. of the 3rd IEEE International Conference on Cluster Computing, pages 372-379. Newport Beach, CA, 2001. Los Alamitos, CA, IEEE Computer Society.
R. V. Dorneles. Particionamento de Domínio e balanceamento de Carga. PhD thesis. Instituto de Informática - UFRGS. Porto Alegre. RS. 2002.
A. Geist et al., PVM: Parallel Virtual Machine. MIT Press. Cambridge. 1994.
I. Ginzburg. AthapascanOb: Intégration Efficace et Portable de Multiprogrammation Légère et de Communication. PhD thesis. Institui National Polytcchnique de Grenoble (INPG). Grenoble. 1997.
GM. Available at http://www.myri.com/scs/. July 2003.
G. Iannello, M. Lauria, and S. Mercolino. Cross-platform analysis of fast messages for myrinet. Lecture Notes in Computer Science. 1362:217-??, 1998.
IEEE. Gigabit ethernet. IEEE P802.3z, 1997.
INSTITUTE OF ELECTRICAL AND ELECTRONIC ENGINEERS. IEEE standard for scalable coherent interface (SCI). IEEE 1596-1992. 1992.
MPI Forum. The MPI message passing interface standard. Technical report. University of Tennesse. Knoxville. Apr. 1994.
MPICH-GM. Avaliable at http://www.myri.com/scs/. July 2003.
L. Prylli and B. Tourancheau. BIP: a new protocol designed for high performance networking on Myrinet. In J. Rolim, editor. Parallel and Distributed Processing - Workshop on Personal Computer Based Networks of Workstations, volume 1388 of Lecture Notes in Computer Science, pages 472-485. Berlin. Springer-Verlag, 1998.
R. L. Rizzi. Modelo Computacional Paralelo para a Hidrodinâmica e para o Transporte de Massa Bidimensional e Multidimensional. PhD thesis. Instituto de Informática - UFRGS. Porto Alegre, RS. 2002.
A. Silveira. R. Ávila. M. Barreto. and P. Navaux. DPC++: Object-oriented programming applied to cluster computing. In H. R. Arabnia, editor. Proc. of the International Conference on Parallel and Distributed Processing Techniques and Applications, volume 5, pages 2515-2521. Las Vegas 2000. Las Vegas, CSREA Press.
TOP500. Available at http://www.top500.org, 2003.
VIA-GM. Available at http://www.myri.com/scs/, July 2003.
M. Barreto, R. Ávila, F. de Oliveira, R. Cassali, and P. Navaux. DECK: an enviroment for parallel programming on clusters of multiprocessors. In Proc. of the 12th Symposium on Computer Architecture and High-Performance Computing, pages 121 - 329. São Pedro, SP, 2000. São Carlos. UFSCAR.
M. Barreto, R. Ávila, and P. Navaux. The MultiCluster model to the integrated use of multiple workstation clusters. In J. Rolim et al., editors. Proc. of the 3rd Workshop on Performance Computer based Networks of Workstations, volume 1800 of Lecture Notes in Computer Science, pages 71-80. Cancun, 2000. Berlin. Springer-Verlag.
M. E. Barreto, P. O. A. Navaux, and M. P. Rivière. DECK: a new model for a distributed executive kernel integrating communication and multithreading for support of distributed object oriented application with fault tolerance support. In Trabajos Seleccionados del 4o. Congreso Argentino de Ciencias de la Computación, volume 2, pages 623-637. Neuquén, Argentina, 1998. Neuquén. Universidad Nacional del Comahue.
N. Boden et al., Myrinet: a gigabit-per-second local-area network. IEEE Micro. 15(1):29-36. Feb. 1995.
F. A. D. de Oliveira. Uma biblioteca para programação paralela por troca de mensagens de clusters baseados na tecnologia SCI. Master's thesis. Instituto de Informática. Universidade Federal do Rio Grande do Sul. Porto Alegre. 2001.
F. A. D. de Oliveira, R. B. A vila, M. E. Barreto, and P. O. A. Navaux. Low-latency and zero-copy message passing protocols for SCI-based clusters. In A. C. M. A. Melo, M. A. R. Dantas, and J. Panetta, editors, Proc. of the 13th Symposium on Computer Architecture and High-Performance Computing, pages 148-155. Pirenópolis, GO, 2001. Brasília. Departamento de Ciência da Computação da UNB.
F. A. D. de Oliveira, R. B. Ávila, M. E. Barreto, P. O. A. Navaux, and C. De Rose. DECK-SCI: High-performance communication and multithreading for SCI clusters. In D. S. Katz, T. Sterling, M. Baker, L. Bergman, M. Paprzyeki, and R. Buyya, editors. Proc. of the 3rd IEEE International Conference on Cluster Computing, pages 372-379. Newport Beach, CA, 2001. Los Alamitos, CA, IEEE Computer Society.
R. V. Dorneles. Particionamento de Domínio e balanceamento de Carga. PhD thesis. Instituto de Informática - UFRGS. Porto Alegre. RS. 2002.
A. Geist et al., PVM: Parallel Virtual Machine. MIT Press. Cambridge. 1994.
I. Ginzburg. AthapascanOb: Intégration Efficace et Portable de Multiprogrammation Légère et de Communication. PhD thesis. Institui National Polytcchnique de Grenoble (INPG). Grenoble. 1997.
GM. Available at http://www.myri.com/scs/. July 2003.
G. Iannello, M. Lauria, and S. Mercolino. Cross-platform analysis of fast messages for myrinet. Lecture Notes in Computer Science. 1362:217-??, 1998.
IEEE. Gigabit ethernet. IEEE P802.3z, 1997.
INSTITUTE OF ELECTRICAL AND ELECTRONIC ENGINEERS. IEEE standard for scalable coherent interface (SCI). IEEE 1596-1992. 1992.
MPI Forum. The MPI message passing interface standard. Technical report. University of Tennesse. Knoxville. Apr. 1994.
MPICH-GM. Avaliable at http://www.myri.com/scs/. July 2003.
L. Prylli and B. Tourancheau. BIP: a new protocol designed for high performance networking on Myrinet. In J. Rolim, editor. Parallel and Distributed Processing - Workshop on Personal Computer Based Networks of Workstations, volume 1388 of Lecture Notes in Computer Science, pages 472-485. Berlin. Springer-Verlag, 1998.
R. L. Rizzi. Modelo Computacional Paralelo para a Hidrodinâmica e para o Transporte de Massa Bidimensional e Multidimensional. PhD thesis. Instituto de Informática - UFRGS. Porto Alegre, RS. 2002.
A. Silveira. R. Ávila. M. Barreto. and P. Navaux. DPC++: Object-oriented programming applied to cluster computing. In H. R. Arabnia, editor. Proc. of the International Conference on Parallel and Distributed Processing Techniques and Applications, volume 5, pages 2515-2521. Las Vegas 2000. Las Vegas, CSREA Press.
TOP500. Available at http://www.top500.org, 2003.
VIA-GM. Available at http://www.myri.com/scs/, July 2003.
Publicado
10/11/2003
Como Citar
MARQUEZAN, Clarissa Cassales; ÁVILA, Rafael Bohrer; NAVAUX, Philippe O. A..
DECK-GM: Uma Implementação do Ambiente DECK para Myrinet. In: SIMPÓSIO EM SISTEMAS COMPUTACIONAIS DE ALTO DESEMPENHO (SSCAD), 4. , 2003, São Paulo.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2003
.
p. 1-8.
DOI: https://doi.org/10.5753/wscad.2003.19096.
