Write Anywhere File Layout

From Wikipedia, the free encyclopedia

WAFL
Developer	NetApp
Full name	Write Anywhere File Layout
Introduced	()
Partition identifier
Structures
Directory contents
File allocation
Bad blocks
Limits
Max file size
Max number of files
Max filename length
Max volume size	16Tb
Allowed characters in filenames
Features
Dates recorded
Date range
Date resolution
Forks
Attributes
File system permissions
Transparent compression
Transparent encryption
Supported operating systems

The Write Anywhere File Layout (WAFL) is a file system that supports large, high-performance RAID arrays, quick restarts without lengthy consistency checks in the event of a crash or power failure, and growing the filesystem size quickly. It was designed by NetApp for use in its storage appliances.

1 Features
2 See also
3 Notes
4 External links

[edit] Features

One of WAFL's most salient features is the snapshot, or read-only copy, of the file system. Snapshots allow users to recover files that they have accidentally deleted; they provide an online backup that can be accessed quickly. It is implemented similarly to that of a log-structured file system. A special kind of snapshot that the filer uses internally called a consistency point allows WAFL to restart quickly in the event of an improper shutdown. NetApp's Data ONTAP Release 7G operating system supports a read-write snapshot called FlexClone.

An important feature of WAFL is its support for both a Unix-style file and directory model for NFS clients and a Microsoft Windows-style file and directory model for CIFS clients. WAFL also supports both security models, including a mode where both security models can be applied to a single file. Unix can use either^[1] access control lists (ACL) or a simple bitmask, whereas the more recent Windows model is based on access control lists. These two features make it possible to write a file to a CIFS type of networked filesystem and access it later via NFS from a Unix workstation.

As the name suggests, Write Anywhere File Layout automatically fragments data using temporal locality to write metadata alongside user data. This fragmentation does not adversely affect files that are sequentially written to or randomly read from, but does affect sequential read after random write. Data ONTAP has the reallocate command as of 7G to perform scheduled and manual defragmentation. Prior to 7G, the wafl scan reallocate command would need to be invoked from an advanced privilege level and could not be scheduled.