Brief Announcement: (Near) Zero-Overhead C++ Bindings for MPI

Abstract:

The Message-Passing Interface (MPI) and C++ form the backbone of high-performance computing and algorithmic research in the field of distributed-memory computing, but MPI only provides C and Fortran bindings. This provides good language interoperability, but higher-level programming languages make development quicker and less error-prone. We propose novel C++ language bindings designed to cover the whole range of abstraction levels from low-level MPI calls to convenient STL-style bindings, where most parameters are inferred from a small subset of the full parameter set. This allows for both rapid prototyping and fine-tuning of distributed code with predictable runtime behavior and memory management. Using template-metaprogramming, only code paths required for computing missing parameters are generated at compile time, which results in (near) zero-overhead bindings.

SEEK ID: https://publications.h-its.org/publications/1903

DOI: 10.1145/3626183.3660260

Research Groups: Computational Molecular Evolution

Publication type: Proceedings

Journal: Proceedings of the 36th ACM Symposium on Parallelism in Algorithms and Architectures

Book Title: Proceedings of the 36th ACM Symposium on Parallelism in Algorithms and Architectures

Publisher: ACM

Citation: Proceedings of the 36th ACM Symposium on Parallelism in Algorithms and Architectures,pp.289-291,ACM

Date Published: 17th Jun 2024

Registered Mode: by DOI

Authors: Kunal Agrawal, Erez Petrank, Demian Hespe, Lukas Hübner, Florian Kurpicz, Peter Sanders, Matthias Schimek, Daniel Seemaier, Tim Niklas Uhl

help Submitter
Citation
Hespe, D., Hübner, L., Kurpicz, F., Sanders, P., Schimek, M., Seemaier, D., & Uhl, T. N. (2024). Brief Announcement: (Near) Zero-Overhead C++ Bindings for MPI. In Proceedings of the 36th ACM Symposium on Parallelism in Algorithms and Architectures (pp. 289–291). ACM. https://doi.org/10.1145/3626183.3660260
Activity

Views: 4

Created: 9th Jan 2025 at 10:19

Last updated: 9th Jan 2025 at 10:20

help Tags

This item has not yet been tagged.

help Attributions

None

Powered by
(v.1.15.2)
Copyright © 2008 - 2024 The University of Manchester and HITS gGmbH