Lustre Unveiled: Evolution, Design, Advancements, and Current Trends

Anjus George; Andreas Dilger; Michael J. Brim; Richard Mohr; Amir Shehata; Jong Youl Choi; Ahmad Maroof Karimi; Jesse Hanley; James Simmons; Dominic Manno; Veronica Melesse Vergara; Sarp Oral; Christopher Zimmer

doi:10.1145/3736583

Lustre Unveiled: Evolution, Design, Advancements, and Current Trends

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

The Lustre filesystem serves as a vital element in high-performance parallel storage, meeting the rising demands of scientific, research, and enterprise environments. Widely deployed across HPC environments, ranging from small-scale applications in AI/ML, to domains like oil and gas, drug discovery, and meteorology, and manufacturing, Lustre addresses the universal challenge of efficiently accessing vast and ever-increasing volumes of data. Lustre is the filesystem of choice on six out of the top 10 fastest supercomputers in the world today, over 65% of the top 100, and also for over 60% of the top 500. Despite its widespread popularity, there is a lack of a complete and up-to-date reference, covering Lustre’s evolution, design, and various advancements made over the years. In this journal, we aim to fill this gap by providing a comprehensive journey of Lustre, including its history with significant contributions to HPC, detailed architecture and design elements, exploration of advancements added through its evolution, and future directions. Additionally, we present a comparison of Lustre with other prominent storage technologies of the era. To illustrate the current state of Lustre, we analyze several filesystem trends, including utilization, performance, and usage patterns on Orion, the Lustre filesystem on the first exascale supercomputer Frontier. We hope that this journal serves as a comprehensive educational reference for the current and future generations interested in HPC filesystem storage aspects.

Original language	English
Article number	21
Journal	ACM Transactions on Storage
Volume	21
Issue number	3
DOIs	https://doi.org/10.1145/3736583
State	Published - Jun 18 2025

Funding

We thank our colleagues from the National Center for Computational Sciences at Oak Ridge National Laboratory, especially Dustin Leverman for providing access to the Orion filesystem performance evaluations. We acknowledge the “Lustre Software Release 2.x Operations Manual”, that assisted us in writing few sections related to resilience and availability aspects of the paper. We also extend acknowledgement to the previously published studies on Lustre (Lustre internals technical report, Lustre white papers by Lustre authors, wiki.whamcloud, and wiki.lustre documentation pages) as referenced in our paper for enhancing our understanding and gathering information on various concepts presented in the paper. We thank the anonymous reviewers and the journal editors for their valuable comments and suggestions, which have significantly improved our manuscript. This research was sponsored by and used resources of the Oak Ridge Leadership Computing Facility (OLCF), which is a DOE Office of Science User Facility at the Oak Ridge National Laboratory supported by the U.S. Department of Energy under Contract No. DE-AC05-00OR22725.

Keywords

High Performance Computing (HPC)
Parallel file system
frontier exascale supercomputer
orion filesystem

Access to Document

10.1145/3736583

Cite this

@article{18f0aad9b2504873bb58a2eff1970bcf,

title = "Lustre Unveiled: Evolution, Design, Advancements, and Current Trends",

abstract = "The Lustre filesystem serves as a vital element in high-performance parallel storage, meeting the rising demands of scientific, research, and enterprise environments. Widely deployed across HPC environments, ranging from small-scale applications in AI/ML, to domains like oil and gas, drug discovery, and meteorology, and manufacturing, Lustre addresses the universal challenge of efficiently accessing vast and ever-increasing volumes of data. Lustre is the filesystem of choice on six out of the top 10 fastest supercomputers in the world today, over 65\% of the top 100, and also for over 60\% of the top 500. Despite its widespread popularity, there is a lack of a complete and up-to-date reference, covering Lustre{\textquoteright}s evolution, design, and various advancements made over the years. In this journal, we aim to fill this gap by providing a comprehensive journey of Lustre, including its history with significant contributions to HPC, detailed architecture and design elements, exploration of advancements added through its evolution, and future directions. Additionally, we present a comparison of Lustre with other prominent storage technologies of the era. To illustrate the current state of Lustre, we analyze several filesystem trends, including utilization, performance, and usage patterns on Orion, the Lustre filesystem on the first exascale supercomputer Frontier. We hope that this journal serves as a comprehensive educational reference for the current and future generations interested in HPC filesystem storage aspects.",

keywords = "High Performance Computing (HPC), Parallel file system, frontier exascale supercomputer, orion filesystem",

author = "Anjus George and Andreas Dilger and Brim, \{Michael J.\} and Richard Mohr and Amir Shehata and Choi, \{Jong Youl\} and Karimi, \{Ahmad Maroof\} and Jesse Hanley and James Simmons and Dominic Manno and Vergara, \{Veronica Melesse\} and Sarp Oral and Christopher Zimmer",

note = "Publisher Copyright: {\textcopyright} 2025 Copyright held by the owner/author(s).",

year = "2025",

month = jun,

day = "18",

doi = "10.1145/3736583",

language = "English",

volume = "21",

journal = "ACM Transactions on Storage",

issn = "1553-3077",

publisher = "Association for Computing Machinery (ACM)",

number = "3",

}

TY - JOUR

T1 - Lustre Unveiled

T2 - Evolution, Design, Advancements, and Current Trends

AU - George, Anjus

AU - Dilger, Andreas

AU - Brim, Michael J.

AU - Mohr, Richard

AU - Shehata, Amir

AU - Choi, Jong Youl

AU - Karimi, Ahmad Maroof

AU - Hanley, Jesse

AU - Simmons, James

AU - Manno, Dominic

AU - Vergara, Veronica Melesse

AU - Oral, Sarp

AU - Zimmer, Christopher

PY - 2025/6/18

Y1 - 2025/6/18

N2 - The Lustre filesystem serves as a vital element in high-performance parallel storage, meeting the rising demands of scientific, research, and enterprise environments. Widely deployed across HPC environments, ranging from small-scale applications in AI/ML, to domains like oil and gas, drug discovery, and meteorology, and manufacturing, Lustre addresses the universal challenge of efficiently accessing vast and ever-increasing volumes of data. Lustre is the filesystem of choice on six out of the top 10 fastest supercomputers in the world today, over 65% of the top 100, and also for over 60% of the top 500. Despite its widespread popularity, there is a lack of a complete and up-to-date reference, covering Lustre’s evolution, design, and various advancements made over the years. In this journal, we aim to fill this gap by providing a comprehensive journey of Lustre, including its history with significant contributions to HPC, detailed architecture and design elements, exploration of advancements added through its evolution, and future directions. Additionally, we present a comparison of Lustre with other prominent storage technologies of the era. To illustrate the current state of Lustre, we analyze several filesystem trends, including utilization, performance, and usage patterns on Orion, the Lustre filesystem on the first exascale supercomputer Frontier. We hope that this journal serves as a comprehensive educational reference for the current and future generations interested in HPC filesystem storage aspects.

AB - The Lustre filesystem serves as a vital element in high-performance parallel storage, meeting the rising demands of scientific, research, and enterprise environments. Widely deployed across HPC environments, ranging from small-scale applications in AI/ML, to domains like oil and gas, drug discovery, and meteorology, and manufacturing, Lustre addresses the universal challenge of efficiently accessing vast and ever-increasing volumes of data. Lustre is the filesystem of choice on six out of the top 10 fastest supercomputers in the world today, over 65% of the top 100, and also for over 60% of the top 500. Despite its widespread popularity, there is a lack of a complete and up-to-date reference, covering Lustre’s evolution, design, and various advancements made over the years. In this journal, we aim to fill this gap by providing a comprehensive journey of Lustre, including its history with significant contributions to HPC, detailed architecture and design elements, exploration of advancements added through its evolution, and future directions. Additionally, we present a comparison of Lustre with other prominent storage technologies of the era. To illustrate the current state of Lustre, we analyze several filesystem trends, including utilization, performance, and usage patterns on Orion, the Lustre filesystem on the first exascale supercomputer Frontier. We hope that this journal serves as a comprehensive educational reference for the current and future generations interested in HPC filesystem storage aspects.

KW - High Performance Computing (HPC)

KW - Parallel file system

KW - frontier exascale supercomputer

KW - orion filesystem

UR - https://www.scopus.com/pages/publications/105012312409

U2 - 10.1145/3736583

DO - 10.1145/3736583

M3 - Article

AN - SCOPUS:105012312409

SN - 1553-3077

VL - 21

JO - ACM Transactions on Storage

JF - ACM Transactions on Storage

IS - 3

M1 - 21

ER -

Lustre Unveiled: Evolution, Design, Advancements, and Current Trends

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this