Results 271 to 280 of about 278,558 (314)

Characterization of MPI Usage on a Production Supercomputer

International Conference for High Performance Computing, Networking, Storage and Analysis, 2018
MPI is the most prominent programming model used in scientific computing today. Despite the importance of MPI, however, how scientific computing applications use it in production is not well understood.
Sudheer Chunduri, Scott Parker, P. Balaji, K. Harms, Kalyan Kumaran   +4 more
semanticscholar   +1 more source

Impacts of Multi-GPU MPI Collective Communications on Large FFT Computation

Workshop on Exascale MPI, 2019
Most applications targeting exascale, such as those part of the Exascale Computing Project (ECP), are designed for heterogeneous architectures and rely on the Message Passing Interface (MPI) as their underlying parallel programming model.
Alan Ayala, S. Tomov, Xi Luo, Hejer Shaeik, A. Haidar, G. Bosilca, J. Dongarra   +6 more
semanticscholar   +1 more source

Open MPI: A High-Performance, Heterogeneous MPI

2006 IEEE International Conference on Cluster Computing, 2006
The growth in the number of generally available, distributed, heterogeneous computing systems places increasing importance on the development of user-friendly tools that enable application developers to efficiently use these resources. Open MPI provides support for several aspects of heterogeneity within a single, open-source MPI implementation ...
Richard Graham, Galen Shipman, Brian Barrett, Ralph Castain, George Bosilca, Andrew Lumsdaine   +5 more
openaire   +1 more source

MPI

2008
?????????????????? ?????????? ???????????????????????????? ?? ?????????????? MPI-????????????????, ?????????????? ???????????????????????? ???? ???????????????????????????????????????????? ??????????????. ?????????????????????? ???????????????? ???????????????????? ?????????????? ???????????????????????????? ???????????? ?????? ?????????????????????????
openaire   +3 more sources

MPI-Mitten: Enabling Migration Technology in MPI

Sixth IEEE International Symposium on Cluster Computing and the Grid (CCGRID'06), 2006
Group communications are commonly used in parallel and distributed environment. However, existing migration mechanisms do not support group communications. This weakness prevents migrationbased proactive fault tolerance, among others, to be applied to MPI applications.
null Cong Du, null Xian-He Sun Sun
openaire   +1 more source

Dyn-MPI

Proceedings of the 2003 ACM/IEEE conference on Supercomputing, 2003
Distributing data is a fundamental problem in implementing efficient distributed-memory parallel programs. The problem becomes more difficult in environments where the participating nodes are not dedicated to a parallel application. We are investigating the data distribution problem in non dedicated environments in the context of explicit message ...
D. Brent Weatherly, David K. Lowenthal, Mario Nakazawa, Franklin Lowenthal   +3 more
openaire   +1 more source

Design analysis of an MPI human functional brain scanner

International journal on magnetic particle imaging, 2017
MPI’s high sensitivity makes it a promising modality for imaging brain function. Functional contrast is proposed based on blood SPION concentration changes due to Cerebral Blood Volume (CBV) increases during activation, a mechanism utilized in fMRI ...
E. Mason, C. Cooley, S. Cauley, M. Griswold, S. Conolly, L. Wald   +5 more
semanticscholar   +1 more source

BCS-MPI

Proceedings of the 2003 ACM/IEEE conference on Supercomputing, 2003
Buffered CoScheduled MPI (BCS-MPI) introduces a new approach to design the communication layer for large-scale parallel machines. The emphasis of BCS-MPI is on the global coordination of a large number of communicating processes rather than on the traditional optimization of the point-to-point performance.
Fernández, J. C. (Juan C.), Petrini, F. (Fabrizio), Frachtenberg, E. (Eitan)   +2 more
openaire   +1 more source

Optimized Broadcast for Deep Learning Workloads on Dense-GPU InfiniBand Clusters: MPI or NCCL?

European MPI Users Group Meeting, 2017
Traditionally, MPI runtimes have been designed for clusters with a large number of nodes. However, with the advent of MPI+CUDA applications and dense multi-GPU systems, it has become important to design efficient communication schemes.
A. A. Awan, Ching-Hsiang Chu, H. Subramoni, D. Panda   +3 more
semanticscholar   +1 more source

Advanced MPI Including New MPI-3 Features

2012
This tutorial will cover several advanced topics in MPI. We will cover one-sided communication, dynamic processes, multithreaded communication and hybrid programming, and parallel I/O. We will also discuss new features in the newest version of MPI, MPI-3, which is expected to be officially released a few days before this tutorial.
William Gropp, Ewing Lusk, Rajeev Thakur
openaire   +1 more source