COSMIC: Enabling Full-Stack Co-Design and Optimization of Distributed Machine Learning Systems

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Raju, Aditi, Ni, Jared, Won, William, Man, Changhai, Krishnan, Srivatsan, Sridharan, Srinivas, Yazdanbakhsh, Amir, Krishna, Tushar, Reddi, Vijay Janapa
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918027974410240
author Raju, Aditi
Ni, Jared
Won, William
Man, Changhai
Krishnan, Srivatsan
Sridharan, Srinivas
Yazdanbakhsh, Amir
Krishna, Tushar
Reddi, Vijay Janapa
author_facet Raju, Aditi
Ni, Jared
Won, William
Man, Changhai
Krishnan, Srivatsan
Sridharan, Srinivas
Yazdanbakhsh, Amir
Krishna, Tushar
Reddi, Vijay Janapa
contents Large-scale machine learning models necessitate distributed systems, posing significant design challenges due to the large parameter space across distinct design stacks. Existing studies often focus on optimizing individual system aspects in isolation. This work challenges this limitation and introduces COSMIC, a full-stack distributed machine learning systems environment enabling end-to-end simulation and agent-based design space exploration. To facilitate efficient exploration and optimization across the entire stack, we introduce Parameter Set Architecture-an abstraction concept analogous to the instruction set architecture-abstracting away configuration complexities of agent-based search methods. Case studies demonstrate COSMIC's ability to consolidate parameters across multiple layers of design abstraction, discovering eight non-obvious high-performance system configurations across four transformer-based models with up to 175 billion parameters. By optimizing across the stack, COSMIC full-stack optimization delivers 1.50-48.41x higher performance compared to the isolated single-stack optimization.
format Preprint
id arxiv_https___arxiv_org_abs_2505_15020
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle COSMIC: Enabling Full-Stack Co-Design and Optimization of Distributed Machine Learning Systems
Raju, Aditi
Ni, Jared
Won, William
Man, Changhai
Krishnan, Srivatsan
Sridharan, Srinivas
Yazdanbakhsh, Amir
Krishna, Tushar
Reddi, Vijay Janapa
Distributed, Parallel, and Cluster Computing
Large-scale machine learning models necessitate distributed systems, posing significant design challenges due to the large parameter space across distinct design stacks. Existing studies often focus on optimizing individual system aspects in isolation. This work challenges this limitation and introduces COSMIC, a full-stack distributed machine learning systems environment enabling end-to-end simulation and agent-based design space exploration. To facilitate efficient exploration and optimization across the entire stack, we introduce Parameter Set Architecture-an abstraction concept analogous to the instruction set architecture-abstracting away configuration complexities of agent-based search methods. Case studies demonstrate COSMIC's ability to consolidate parameters across multiple layers of design abstraction, discovering eight non-obvious high-performance system configurations across four transformer-based models with up to 175 billion parameters. By optimizing across the stack, COSMIC full-stack optimization delivers 1.50-48.41x higher performance compared to the isolated single-stack optimization.
title COSMIC: Enabling Full-Stack Co-Design and Optimization of Distributed Machine Learning Systems
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2505.15020