Flint: Compiler Enabled Cluster-Free Design Space Exploration for Distributed ML

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Yoo, Jinsun, Cowan, Meghan, Du, Zheng, Man, Changhai, Sridharan, Srinivas, Krishna, Tushar
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913045426470912
author Yoo, Jinsun
Cowan, Meghan
Du, Zheng
Man, Changhai
Sridharan, Srinivas
Krishna, Tushar
author_facet Yoo, Jinsun
Cowan, Meghan
Du, Zheng
Man, Changhai
Sridharan, Srinivas
Krishna, Tushar
contents Design space exploration for future distributed Machine Learning systems suffers from a lack of readily available workload representation that enables flexible exploration across the stack. We present Flint, a framework that bridges this gap by leveraging the Intermediate Representation of Machine Learning framework compilers. The compiler does the heavy weight lifting of understanding and preserving the behavior of the original model code. Flint can collect the workload representation of arbitrary cluster size because it interfaces with the compiler before hardware execution. We validate the workload graph against post-execution traces and show the flexibility of Flint through a design space exploration case study.
format Preprint
id arxiv_https___arxiv_org_abs_2604_17550
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Flint: Compiler Enabled Cluster-Free Design Space Exploration for Distributed ML
Yoo, Jinsun
Cowan, Meghan
Du, Zheng
Man, Changhai
Sridharan, Srinivas
Krishna, Tushar
Distributed, Parallel, and Cluster Computing
Design space exploration for future distributed Machine Learning systems suffers from a lack of readily available workload representation that enables flexible exploration across the stack. We present Flint, a framework that bridges this gap by leveraging the Intermediate Representation of Machine Learning framework compilers. The compiler does the heavy weight lifting of understanding and preserving the behavior of the original model code. Flint can collect the workload representation of arbitrary cluster size because it interfaces with the compiler before hardware execution. We validate the workload graph against post-execution traces and show the flexibility of Flint through a design space exploration case study.
title Flint: Compiler Enabled Cluster-Free Design Space Exploration for Distributed ML
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2604.17550