Advanced Operating Systems: Lecture 34 - Mr. Farhan Zaidi

Advanced Operating Systems - Lecture 34: Log structured file systems. This lecture will cover the following: log structured file systems; issue - log management; current trend is towards logging FS; linux virtual file system; primary objects in VFS; the sun network file system (NFS);...

CS703 ­ Advanced  Operating Systems

By Mr Farhan Zaidi

34

• Log structured (or journaling) file systems record each update to the file system as a transaction.

• All transactions are written to a log A transaction is considered committed once it is written to the log

However, the file system may not yet be updated.

 Idea: lets keep track of what operations are in progress

and use this for recovery It’s keep a “log” of all operations, upon a crash we can scan through the log and find

problem areas that need fixing

 Add log area to disk

 Always write changes to log first – called write-ahead logging or

journaling.

 Then write the changes to the file system

 All reads go to the file system

 Crash recovery – read log and correct any inconsistencies in the file system

Log File system

 Observation: Log only needed for crash recovery

 Checkpoint operation – make in-memory copy of file system (file cache) consistent with disk

 After a checkpoint, can truncate log and start again.

 Most logging file systems only log metadata (file descriptors and directories) and not file data to keep log size down

 Two disk writes (on different parts of the disk) for every change?

 Synchronous writes are on every file system change?

 Observation: Log writes are sequential on disk so even synchronous writes can be fast

 Best performance if log on separate disk

 Fast recovery: recovery time O(active operations) and not O(disk size)

 Better performance if changes need to be reliable

 If you need to do synchronous writes, sequential

synchronous writes are much faster than non-sequential ones.

 Examples:

 Windows NTFS

 Veritas on Sun

 Many competing logging file system for Linux

 Uniform file system interface to user processes

 Represents any conceivable file system’s general

feature and behavior

 Assumes files are objects that share basic properties regardless of the target file system

 Superblock object

 Represents a specific mounted file system

 Inode object

 Represents a specific file

 Dentry object

 Represents a specific directory entry

 File object

 Represents an open file associated with a process

 An implementation and a specification of a software system for accessing remote files across LANs (or WANs)

 The implementation is part of the Solaris and SunOS

operating systems running on Sun workstations using an

unreliable datagram protocol (UDP/IP protocol and Ethernet

Schematic View of NFS Architecture