Work

Improving the Scaling of an Asynchronous Many-Task Runtime with a Lightweight Communication Engine

Year: 2023

Type: conference-paper

Abstract: There is a growing interest in Asynchronous Many-Task (AMT) runtimes as an efficient way to map irregular and dynamic parallel applications onto heterogeneous computing resources. In this work, we show that AMTs nonetheless struggle with communication bottlenecks when scaling computations strongly and that the design of commonly-used communication libraries such as MPI contribute to these bottlenecks. We replace MPI with LCI, a Lightweight Communication Interface that is designed for dynamic, asynchronous frameworks, as the communication layer for the PaRSEC runtime. The result is a significant reduction of end-to-end latency in communication microbenchmarks and a reduction of overall time-to-solution by up to 12% in HiCMA, a tile-based low-rank Cholesky factorization package.

Source: Proceedings of the International Conference on Parallel Processing

DOI: 10.1145/3605573.3605642

Authors: Omri Mor, George Bosilca, Marc Snir

Institutions: University of Illinois Urbana-Champaign, University of Illinois System, University of Tennessee at Knoxville

Language: English

Topic: Parallel Computing and Optimization Techniques

Subfield: Hardware and Architecture

Field: Computer Science

Domain: Physical Sciences

Open Access status: bronze

Funder: National Science Foundation

Awards: OAC Core: Small: Collaborative Research: Scalable Run-Time for Highly Parallel, Heterogeneous Systems, OAC Core: Small: Collaborative Research: Scalable Run-Time for Highly Parallel, Heterogeneous Systems