By Topic

The performance analysis of portable parallel programming interface MpC for SDSM and pthread

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

1 Author(s)
Midorikawa, H. ; Dept. of Comput. & Inf. Sci., Seikei Univ., Tokyo, Japan

In parallel programming model, MPI seems to have established its position, and OpenMP is extensively investigated as the next standard. However, OpenMP is not so efficient for clusters. Using OpenMP on clusters causes more performance degradation than using SDSMs directly, because most of the OpenMP implementations for clusters use SDSMs in their under layer. This paper presents the performance evaluation of new portable parallel programming interface MpC, Meta process C. It is a minimal extension of ANSI C and its API also becomes a universal API for various SDSMs and pthread. The MpC is based on meta process model. The meta process model is a parallel programming paradigm based on a hierarchical shared memory model and an explicit description of parallelism. The model introduces 'shared' data that can be accessed by all processes in one meta process and distinguishes process-local and process-shared data explicitly. Using a hierarchical data scope, the process-local data are practically prohibited to be accessed by other processes. The paper summarizes the model feature, and compares performance and productivity of MpC with other languages, OpenMP and UPC. It also proves good portability of MpC programs for clusters and shared memory machines.

Published in:

Cluster Computing and the Grid, 2005. CCGrid 2005. IEEE International Symposium on  (Volume:2 )

Date of Conference:

9-12 May 2005