A Novel Sequential Tree Algorithm Based on Scoreboard for MPI Broadcast Communication

Won-young CHUNG  Jae-won PARK  Seung-Woo LEE  Won Woo RO  Yong-surk LEE  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E94-D   No.12   pp.2523-2527
Publication Date: 2011/12/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.E94.D.2523
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Computer System
Keyword: 
MPI,  broadcast communication,  multicore,  distributed memory,  MPSoC,  

Full Text: PDF(846.8KB)
>>Buy this Article


Summary: 
The message passing interface (MPI) broadcast communication commonly causes a severe performance bottleneck in multicore system that uses distributed memory. Thus, in this paper, we propose a novel algorithm and hardware structure for the MPI broadcast communication to reduce the bottleneck situation. The transmission order is set based on the state of each processing node that comprises the multicore system, so the novel algorithm minimizes the performance degradation caused by conflict. The proposed scoreboard MPI unit is evaluated by modeling it with SystemC and implemented using VerilogHDL. The size of the proposed scoreboard MPI unit occupies less than 1.03% of the whole chip, and it yields a highly improved performance up to 75.48% as its maximum with 16 processing nodes. Hence, with respect to low-cost design and scalability, this scoreboard MPI unit is particularly useful towards increasing overall performance of the embedded MPSoC.