The allreduce collective operation accounts for a significant fraction of the runtime of
workloads running on distributed systems. One factor determining its performance is the …

Applications often communicate data that is non-contiguous in the send-or the receive-
buffer, eg, when exchanging a column of a matrix stored in row-major order. While non …

An increasing number of MPI applications are being ported to take advantage of the
compute power offered by GPUs. Data movement continues to be the major bottleneck on …

This paper addresses performance-portability and overall performance issues when derived
datatypes are used with four MPI implementations: Open MPI, MPICH, MVAPICH2, and Intel …

Derived datatypes are commonly used in MPI applications to exchange non-contiguous data
among processes. However, state-of-the-art MPI libraries do not offer efficient processing of …

This paper addresses the challenges of MPI derived datatype processing and proposes
FALCON-X—A Fast and Low-overhead Communication framework for optimized zero-copy …

We examine natural expectations on communication performance using MPI derived
datatypes in comparison to the baseline," raw" performance of communicating simple …

Noncontiguous data communication has been heavily adopted in scientific applications,
especially for those written with MPI. Common strategies to handle noncontiguous data, like …

In the era of post-Moore computing, network offload emerges as a solution to two
challenges: the imperative for low-latency communication and the push towards hardware …

Data redistribution aims to reshuffle data to optimize some objective for an algorithm. The
objective can be multi-dimensional, such as improving computational load balance or …