Lustre File System

chris

Outlines
   

What is lustre How does it works Features Performance

What is Lustre?

Cluster File Systems, Inc. (CFS) is the company that originally developed the Lustre distributed file system. On September 12, 2007, Sun Microsystems, Inc. signed an agreement to acquire CFS.[1] The acquisition was completed on October 1, 2007. Lustre is a shared disk file system, generally used for large scale cluster computing. The name Lustre is a portmanteau of Linux and cluster. The project aims to provide a file system for clusters of tens of thousands of nodes with petabytes of storage capacity, without compromising speed or security. Lustre is available under the GNU GPL.

Support Platform

Lustre Architecture

Lustre Components

MGS  The Management Server (MGS) defines configuration information for all Lustre

filesystems at a site.

Each Lustre target contacts the MGS to provide information, and Lustre clients contact the MGS to retrieve information. The MGS can provide live updates to the configuration of targets and clients. The MGS requires its own disk for storage. However, there is a provision that allows the MGS to share a disk ("co-locate") with a single MDT.

Lustre Components

MDT  The MDT provides back-end storage for metadata for a single filesystem. The

Metadata Server (MDS) provides the network request handling for one or more local MDTs.
The metdata managed by the MDT consists of the file hierarchy ("namespace"), along with file attributes such as permissions and references to the data objects stored on the OSTs.

Lustre Components

OSS and OSTs

An OST provides back-end storage for file object data (effectively, chunks of user files). Typically, multiple OSTs provide access to different file chunks. The MDT tracks the location of the chunks. On a node serving OSTs, an Object Storage Server (OSS) component provides the network request handling for one or more local OSTs.

Lustre Configuration
1.

 

Create Lustre format file system on MGS/OSS
mkfs.lustre
--fsname={whatever} --{--ost/--mdt/--mgs}
--mgsnode={hostname@tcp0}

/dev/sda

2.
 

Mount lustre filesystem on MGS and OSS
mount –t lustre /dev/sda /mnt/test mount mgs/mdt at first

3.

Mount lustre filesystem on clients
mount –t lustre {hostname@tcp0}:/{fsname} /mnt/client_FS

Lustre Configuration

Lustre Failover mechanism
   

Active/active Active/passive Heartbeat STONITH (Shoot The Other Node In The Head

Heartbeat OSS-1 STONITH poweroff

msg

OSS1backu p

Lustre Failover mechanism

In Lustre, a disk failure is an unrecoverable error. For this reason, you must have reliable back-end storage with RAID. Backup device

dd if=/dev/{original} of=/dev/{new} bs=4k conv=sync,noerror

Features
      

Kerberos 、 ACL Bonding Live Add/Remove OSTs SNMP Quota File Stripe size Using POSIX Standard

Performance

Read/write files with different kernel io scheduler

Performance

Create files with different Kernel io scheduler

Performance

Multi-clients

Performance

Lustre vs GPFS

Reference
1. 2. 3.

Sun Microsystems Expands High Performance Co , press release, September 12, 2007. Lustre™ 1.6 Operations Manual http://wiki.lustre.org/index.php? title=Main_Page

Any Questions?

Sign up to vote on this title
UsefulNot useful