MPI performance engineering with the MPI tool interface: the integration of MVAPICH and TAU
MPI implementations are becoming increasingly complex and highly tunable, and thus
scalability limitations can come from numerous sources. The MPI Tools Interface (MPI_T) …
scalability limitations can come from numerous sources. The MPI Tools Interface (MPI_T) …
Designing a profiling and visualization tool for scalable and in-depth analysis of high-performance GPU clusters
P Kousha, B Ramesh, KK Suresh… - 2019 IEEE 26th …, 2019 - ieeexplore.ieee.org
The recent advent of advanced fabrics like NVIDIA NVLink is enabling the deployment of
dense Graphics Processing Unit (GPU) systems, eg, DGX-2 and Summit. The Message …
dense Graphics Processing Unit (GPU) systems, eg, DGX-2 and Summit. The Message …
INAM2: InfiniBand Network Analysis and Monitoring with MPI
H Subramoni, AM Augustine, M Arnold… - … Conference on High …, 2016 - Springer
Modern high-end computing is being driven by the tight integration of several hardware and
software components. On the hardware front, there are the multi-/many-core architectures …
software components. On the hardware front, there are the multi-/many-core architectures …
A survey of methods for collective communication optimization and tuning
U Wickramasinghe, A Lumsdaine - arXiv preprint arXiv:1611.06334, 2016 - arxiv.org
New developments in HPC technology in terms of increasing computing power on
multi/many core processors, high-bandwidth memory/IO subsystems and communication …
multi/many core processors, high-bandwidth memory/IO subsystems and communication …
MR-Advisor: A comprehensive tuning, profiling, and prediction tool for MapReduce execution frameworks on HPC clusters
MapReduce is the most popular parallel computing framework for big data processing which
allows massive scalability across distributed computing environment. Advanced RDMA …
allows massive scalability across distributed computing environment. Advanced RDMA …
Planning for performance: Enhancing achievable performance for MPI through persistent collective operations
Advantages of nonblocking collective communication in MPI have been established over the
past quarter century, even predating MPI-1. For regular computations with fixed …
past quarter century, even predating MPI-1. For regular computations with fixed …
Enabling callback-driven runtime introspection via MPI_T
MA Hermanns, NT Hjlem, M Knobloch… - Proceedings of the 25th …, 2018 - dl.acm.org
Understanding the behavior of parallel applications that use the Message Passing Interface
(MPI) is critical for optimizing communication performance. Performance tools for MPI …
(MPI) is critical for optimizing communication performance. Performance tools for MPI …
Planning for performance: persistent collective operations for MPI
B Morgan, DJ Holmes, A Skjellum… - Proceedings of the 24th …, 2017 - dl.acm.org
Advantages of nonblocking collective communication in MPI have been established over the
past quarter century, even predating MPI-1. For regular computations with fixed …
past quarter century, even predating MPI-1. For regular computations with fixed …
Communication optimization technology based on network dynamic performance model
X Cui, X Li, B Wang - Mathematical Problems in Engineering, 2020 - Wiley Online Library
This work analyses different communication modes in applications of supercomputing,
proposes a communication dynamic performance model based on topology awareness, and …
proposes a communication dynamic performance model based on topology awareness, and …
Sonar: Automated communication characterization for hpc applications
Future computing systems will need to operate within hard power and energy constraints,
this is particularly true for Exascale-class systems. These constraints are hard for technical …
this is particularly true for Exascale-class systems. These constraints are hard for technical …