Impact of interconnection networks and application granularity to compound cluster environments

Network-on-chip architectures are becoming commercial off-the-shelf technology to gather computational resources in cluster environments. However, since legacy of the Ethernet standard is overwhelming on account of numerous devices available in many organizations, aspects of the application such as the ratio of computation and communication should be considered in the design of a computational configuration. In this article, it is presented a study of the impact of interconnection networks to execute applications with different levels of granularity, considering Ethernet and Myrinet interconnection networks.

[1]  Gita Alaghband,et al.  Fundamentals of Parallel Processing , 2002 .

[2]  Jack Dongarra,et al.  Message‐Passing Software Systems , 1999 .

[3]  George Bosilca,et al.  Open MPI: Goals, Concept, and Design of a Next Generation MPI Implementation , 2004, PVM/MPI.

[4]  Peter S. Pacheco.  Parallel programming with MPI , 1996 .

[5]  Mario A. R. Dantas,et al.  Experiences with porting an engineering application onto workstation clusters using PVM and MPI , 1996, EUROSIM.

[6]  Jack Dongarra,et al.  Parallel matrix transpose algorithms on distributed memory concurrent computers , 1993, Proceedings of Scalable Parallel Libraries Conference.

[7]  Wu-chun Feng,et al.  The Quadrics Network: High-Performance Clustering Technology , 2002, IEEE Micro.

[8]  Mario A. R. Dantas.  Evaluation of Process Migration for Parallel Heterogeneous Workstation Clusters , 1998, Euro-Par.

[9]  Alex Rapaport,et al.  Mpi-2: extensions to the message-passing interface , 1997 .

[10]  Anthony Skjellum,et al.  A High-Performance, Portable Implementation of the MPI Message Passing Interface Standard , 1996, Parallel Comput..

[11]  James Arthur Kohl,et al.  PVM and MPI: a Comparison of Features , 2007 .

[12]  Charles L. Seitz,et al.  Myrinet: A Gigabit-per-Second Local Area Network , 1995, IEEE Micro.

[13]  Greg J. Regnier,et al.  The Virtual Interface Architecture , 2002, IEEE Micro.

[14]  Yves Robert,et al.  Revisiting Matrix Product on Master-Worker Platforms , 2006, 2007 IEEE International Parallel and Distributed Processing Symposium.

[15]  Reza Zamani,et al.  Myrinet networks: a performance study , 2004, Third IEEE International Symposium on Network Computing and Applications, 2004. (NCA 2004). Proceedings..

[16]  Scott Rixner,et al.  Comparing Ethernet and Myrinet for MPI communication , 2004, LCR.

[17]  Ian Foster,et al.  Designing and building parallel programs , 1994 .