Testing and Performance Analysis of FCA Algorithm Accelerating IBM Platform MPI

Lin Feng Zhang; Lei Chen

doi:10.4028/www.scientific.net/AMM.539.429

Paper Titles

Design of Employment Management Platform Based on ASP.NET
p.412

Cyclic Structure in Regenerating Codes
p.416

Study of New Materials of Mobile Communication Devices
p.420

The Design of Web User Interface Based on Dynamic Re-Components
p.424

Testing and Performance Analysis of FCA Algorithm Accelerating IBM Platform MPI
p.429

Research on Intelligent Information Search Based on Web
p.434

Design of Information Management System Based on WEB
p.438

Behavior Analysis in E-Commerce
p.443

Architecture and Technique on Internet of Things
p.448

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vol. 539Testing and Performance Analysis of FCA Algorithm...

Testing and Performance Analysis of FCA Algorithm Accelerating IBM Platform MPI

Abstract:

MPI is one of the important standards in high performance computing. MPI performance is generally focused on collective communications. And FCA (Fabric Collective Accelerator) is a new method accelerating collective communications. Through high performance computing environment testing, this paper mainly analyses the result of FCA with shared memory and without share memory accelerating IBM Platform MPI, FCA's principle and integration between IBM Platform MPI and FCA. At the same time, this paper may be a good reference for high performance computing using FCA.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volume 539)

Pages:

429-433

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.539.429

Citation:

Cite this paper

Online since:

July 2014

Authors:

Lin Feng Zhang*, Lei Chen

Keywords:

Collective Communication, FCA, High Performance Computing, IBM Platform MPI

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] Zhihui DU. High-performance Computing Parallel Programming Technology—MPI Parallel Programming Technology [M]. page: 12~15.

Google Scholar

[2] http: /www. mpi-forum. org.

Google Scholar

[3] http: /mvapich. cse. ohio-state. edu.

Google Scholar

[4] http: /www. open-mpi. org.

Google Scholar

[5] http: /www. mcs. anl. gov/research/projects/mpi/mpich1-old.

Google Scholar

[6] Rajeev Thakur, William Gropp, Ewing Lusk. Optimizing Noncontiguous Accesses in MPI-IO.

DOI: 10.1016/s0167-8191(01)00129-6

Google Scholar

[7] http: /mellanox. com/content/pages. php?pg=products_dyn&product_family=104& me nu_section=73.

Google Scholar

[8] Mellanox_FCA V2. 2 User Guider.

Google Scholar

[9] Rajeev Thakur, William Gropp. Improving the Performance of Collective Operations in MPICH.

Google Scholar

[10] Ernie Chan, Robert van de Geijn, William Gropp, Rajeev Thakur. Colletive Communication on Architectures that Support Simultaneous Communication over Mutiple Links.

DOI: 10.1145/1122971.1122975

Google Scholar

[11] Swamy N. Kandadai, Xinghong He. Performance of HPC application over InfiniBand, 10 Gigabit and 1 Gigabit Ethernet.

Google Scholar