2012 | OriginalPaper | Chapter
Design and Implementation of MPI-Based Communication Mechanism in HPC-RTI
Authors : Hongbo Liang, Yiping Yao, Xiaodong Mu, Ling Wang
Published in: Recent Advances in Computer Science and Information Engineering
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
HPC-RTI (High Performance Computing Runtime Infrastructure), the RTI for High-Level Architecture, which executes on high performance computing platform, can achieve better performance than common RTI. In our HPC-RTI, a high speed communication layer based MPI is designed to reduce the communication latency. In this layer, a new process merge method is provided to implement the dynamicMPI process management, and a message sending/receiving mechanism is presented in detail. By this mechanism, message passing latency can be reduced, and network bandwidth can be better utilized. The preliminary experiment results show that this high performance communication layer is well suited for high performance distributed simulation.