arXiv Open Access 2025

PICO: Performance Insights for Collective Operations

Saverio Pasqualoni Lorenzo Piarulli Daniele De Sensi

Lihat Sumber

Abstrak

Collective operations are cornerstones of both HPC application and large-scale AI training and inference. Yet, comprehensive, systematic and reproducible performance evaluation and benchmarking of said operations is not straightforward. Existing frameworks do not provide sufficiently detailed profiling information, nor they ensure reproducibility and extensibility. In this paper, we present PICO (Performance Insights for Collective Operations), a novel lightweight, extensible framework built with the aim of simplifying collective operations benchmarking.

Topik & Kata Kunci

cs.DC cs.PF

Penulis (3)

Saverio Pasqualoni

Lorenzo Piarulli

Daniele De Sensi

Format Sitasi

APA MLA BibTeX

Pasqualoni, S., Piarulli, L., Sensi, D.D. (2025). PICO: Performance Insights for Collective Operations. https://arxiv.org/abs/2508.16809

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2025
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓