Reducing the overhead of intra-node communication in clusters of SMPs

Authors:
Sascha Hunold;Thomas Rauber
Affiliations:
Department of Mathematics, Physics and Computer Science, University of Bayreuth, Germany;Department of Mathematics, Physics and Computer Science, University of Bayreuth, Germany
Venue:
ISPA'05 Proceedings of the Third international conference on Parallel and Distributed Processing and Applications
Year:
2005

Citing 4
Cited 0

A multithreaded message passing interface (MPI) architecture: performance and program issues

Journal of Parallel and Distributed Computing
Optimizing threaded MPI execution on SMP clusters

ICS '01 Proceedings of the 15th international conference on Supercomputing
On the design of Chant: a talking threads package

Proceedings of the 1994 ACM/IEEE conference on Supercomputing
Multilevel hierarchical matrix multiplication on clusters

Proceedings of the 18th annual international conference on Supercomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This article presents the C++ library vShark which reduces the intra-node communication overhead of parallel programs on clusters of SMPs. The library is built on top of message-passing libraries like MPI to provide thread-safe communication but most importantly, to improve the communication between threads within one SMP node. vShark uses a modular but transparent design which makes it independent of specific communication libraries. Thus, different subsystems such as MPI, CORBA, or PVM could also be used for low-level communication. We present an implementation of vShark based on MPI and the POSIX thread library, and show that the efficient intra-node communication of vShark improves the performance of parallel algorithms.