Architecture Descriptions

The architecture descriptions listed below provide information about Lustre architecture and design and are intended to help users better understand the conceptual framework of the Lustre file system.

Note: These documents reflect the state of design of a Lustre feature at a particular point in time. They many contain information that is incomplete or obsolete and may not reflect the current architecture, features and functionality of Lustre.

Adaptive Timeouts - Use Cases

Backup

Caching OSS

Changelogs

Changelogs 1.6

Client Cleanup

Clustered Metadata

Commit on Share

CROW

CTDB with Lustre

Cuts

DMU OSD

DMU Zerocopy

End-to-end Checksumming

Epochs

External File Locking

FIDs on OST

Fileset

Flash Cache

Free Space Management

GNS

HSM

HSM and Cache

HSM Migration

Interoperability fids zfs

Interoperability 1.6 1.8 2.0

IO system

Libcfs

Llog over OSD

LRE Images

Lustre Logging API

MDS striping format

MDS-on-DMU

Metadata API

Migration (1)

Migration (2)

MPI IO and NetCDF

MPI LND

Multiple Interfaces For LNET

Network Request Scheduler

New Metadata API

PAG

Pools of targets

Profiling Tools for IO

Proxy Cache

Punch and Extent Migration

Recovery Failures

Open by fid

Request Redirection

Scalable Pinger

Security

Server Network Striping

Simple Space Balance Migration

Simplified Interoperation

Space Manager

Sub Tree Locks

User Level Access

User Level OSS

Userspace Servers

Version Based Recovery

Wide Striping

Wire Level Protocol

Write Back Cache

ZFS for Lustre

ZFS large dnodes

ZFS TinyZAP