The documents distributed by this server have been provided by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a noncommercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.
Publications of SPCL
|T. Hoefler, P. Gottschling, W. Rehm, A. Lumsdaine:|
|Optimizing a Conjugate Gradient Solver with Non-Blocking Collective Operations|
(Vol , Nr. , In Recent Advantages in Parallel Virtual Machine and Message Passing Interface. 13th European PVM/MPI User's Group Meeting, Proceedings, LNCS 4192, presented in Bonn, Germany, pages 374-382, Springer, ISSN: 0302-9743, ISBN: 3-540-39110-X, Sep. 2006)
Invited to a journal special issue on top picks from EuroMPI'06.
AbstractAbstract. This paper presents a case study about the applicability and usage of non-blocking collective operations. These operations provide the ability to overlap communication with computation and to avoid unnecessary synchronization. We introduce our NBC library, a portable low-overhead implementation of non-blocking collectives on top of MPI-1. We demonstrate the easy usage of the NBC library with the optimization of a conjugate gradient solver with only minor changes to the traditional parallel implementation of the program. The optimized solver runs up to 34% faster and is able to overlap most of the communication. We show that there is, due to the overlap, no performance difference between Gigabit Ethernet and InfiniBandTM for our calculation.