The MPI multithreading model has been historically difficult to optimize; the interface that it
provides for threads was designed as a process-level interface. This model has led to …

Partitioned point-to-point communication primitives provide a performance-oriented
mechanism to support a hybrid parallel programming model and have been included in the …

The Message Passing Interface (MPI) has been one of the most prominent programming
paradigms in high-performance computing (HPC) for the past decade. Lately, with changes …

Partitioned point-to-point communication and persistent collective communication were both
recently standardized in MPI-4.0. Each offers performance and scalability advantages over …

Asynchronous programming models (APM) are gaining more and more traction, allowing
applications to expose the available concurrency to a runtime system tasked with …

Modern HPC systems require efficient hybrid programming model to utilize their hardware
resources effectively. The Message Passing Interface (MPI) has accommodated next …

Disaggregated architecture brings new opportunities to memory-consuming applications like
graph processing. It allows one to outspread memory access pressure from local to far …

The recently proposed MPI Sessions extensions to the MPI standard present a new
paradigm for applications to use with MPI. MPI Sessions has the potential to address several …

The Open MPI for Exascale (OMPI-X) project was one of two in the Exascale Computing
Project (ECP) focused on advancing the MPI ecosystem. The OMPI-X team worked with …

As network speeds increase, the overhead of processing incoming messages is becoming
onerous enough that many manufacturers now provide network interface cards (NICs) with …