Lustre File System - Demo Quick Start Guide

Lustre File System: Demo Quick Start Guide

by Torben Kling-Petersen
March 2009

The Lustreā„¢ file system is a scalable, secure, robust, and highly-available cluster file system that addresses the I/O needs, such as low latency and extreme performance, of large computing clusters. Designed, developed, and maintained by Sun Microsystems, the Lustre file system is intended for environments where traditional shared file systems, such as NFS, do not scale to the required aggregate throughput or large number of nodes.

While the Lustre file system has been around for a number of years in the Open Source arena and there are a large number of installations worldwide, getting started without reading a 500+ page manual - and having to be a Linux expert as well - is difficult. This paper provides a simple cookbook for non-Linux experts on how to set up a Linux-based Lustre file system using small servers, workstations, PCs, or other available hardware for demonstration purposes.

Contents

  • Lustre file system overview
  • Configuration overview
  • Preliminary setup
    • Installing the Linux operating system
    • Creating the virtual volumes
    • Installing the Lustre stack
  • Lustre file system configuration
    • Metadata Server
    • Object Store Servers
    • Client
  • Managing the file system
    • Using stripes
    • Handling full OSTs
    • Migrating data within a file system
  • Summary
  • About the author
  • References
  • Ordering Sun documents
  • Accessing Sun documentation online
About the Authors

Torben Kling-Petersen has worked with high performance computing in one form or another since 1994 and is currently working as a Senior Technical Specialist for HPC in Sun's Lustre Group. Over the years, he has worked in a number of capacities such as lead architect for enterprise datacenter infrastructure, technical research lead and product specialist for high-end visualization, to mention a few. In his present capacity, Torben works in a global role providing technical evangelism and solution architectures on petaflop-scale HPC projects.

Rate this blueprint (Log In to vote.)
Choices Your Vote

Great

Good

Fair

Poor

Labels

new new Delete
blueprints blueprints Delete
blueprint blueprint Delete
Enter labels to add to this page:
Please wait 
Looking for a label? Just start typing.
  1. Jul 17

    lustre_tester says:

    Excellent guide! Thank you much! But there is a small mistake: 7.Login as root...

    Excellent guide!
    Thank you much!

    But there is a small mistake:
    7.Login as root to the MDS, and repeat Steps 1 to 5 to create a single logical volume (MDT) for the MDS. As each file written to the Lustre file system requires 4 KB of storage space in the MDS, a 1 GB logical volume for the MDT is more than sufficient for most typical deployments.

    I think you should add this line:

    "lvcreate --name vg00/mdt --size 1000M"

    And change "logical volume" to "virtual volume" for more clear

    1. Sep 21

      ngrkic says:

      Did you build one volume or six?Was confused building it....

      Did you build one volume or six?Was confused building it....

  2. Sep 21

    ngrkic says:

    I was confused too...And i have one question.Does we need to build one or six ...

    I was confused too...And i have one question.Does we need to build one or six volumes for MDS?

Sign up or Log in to add a comment or watch this page.


The individuals who post here are part of the extended Sun Microsystems community and they might not be employed or in any way formally affiliated with Sun Microsystems. The opinions expressed here are their own, are not necessarily reviewed in advance by anyone but the individual authors, and neither Sun nor any other party necessarily agrees with them.

Copyright 1994-2009 Sun Microsystems, Inc.
Powered by Atlassian Confluence
Sun Guidelines on Public Discourse Privacy Policy Terms of Use Trademarks Site Map Employment Investor Relations Contact