Program

Monday, 13 November 2023

2:00 - 5:30

2:00 - 2:10

Welcome & Introduction

2:10 - 3:00

Distinguished Speaker: Jim Dinan (NVIDIA), "GPU Centric Communication -- Is MPI Missing Out?"

3:00 - 3:30

Coffee Break

3:30 - 3:50

Optimizing Irregular Communication with Neighborhood Collectives and Locality-Aware Parallelism

Gerald Collom, Rui Peng Li, Amanda Bienz

3:50 - 4:10

Using Mixed-Radix Decomposition to Enumerate Computational Resources of Deeply Hierarchical Architectures

Philippe Swartvagher, Sascha Hunold, Jesper Larsson Träff, Ioannis Vardas

4:10 - 4:30

Embedding Rust within Open MPI

Jake Tronge, Howard Pritchard

4:30 - 4:50

A Statistical Analysis of HPC Network Tuning

Donald Kruse, Whit Schonbein, Matthew G.F. Dosanjh

4:50 - 5:10

OpenSHMEM Queues: An abstraction for enhancing message rate, bandwidth utilization, and reducing tail latency in OpenSHMEM Applications

Vishwanath Venkatesan, Manjunath Gorentla Venkata

5:10 - 5:30

Efficient data redistribution for malleable applications

Iker Martín Álvarez, José Aliaga, Maribel Castillo, Sergio Iserte

Page updated

Google Sites

Report abuse