File system

1
File system

• Recitation 8

2
File system

• File system is a logical layer over the device,
  hiding its structure.
• File system provide mechanism for storage and
  access to data and programs.
• File system reside on the secondary storage
  device.
• Must hold large amount of data.
• Data is non-volatile
• Disk drives present performance issues.

3
Physical disk structure

• Disk components
• Platters
• Surfaces
• Tracks
• Sectors
• Cylinders
• Arm
• Logically, disk broken down into sectors
• Addressed by cylinder, head, sector (CHS)

4
Physical disk structure
5
Disk performance

• Disk request performance depends upon a number of
  steps
• Seek moving the disk arm to the correct
  cylinder
• Depends on how fast disk arm can move (increasing
  very slowly)
• Rotation waiting for the sector to rotate under
  the head
• Depends on rotation rate of disk (increasing, but
  slowly)
• Transfer transferring data from surface into
  disk controller
• electronics, sending it back to the host
• Depends on density (increasing quickly)
• When the OS uses the disk, it tries to minimize
  the cost of all of these steps
• Particularly seeks

6
Files

• A file is data with some properties
• Contents, size, owner, last read/write time,
  protection, etc.
• A file can also have a type
• Understood by the file system
• Block, character, device, portal, link, etc.
• Understood by other parts of the OS or runtime
  libraries
• Executable, dll, souce, object, text, etc.
• A files type can be encoded in its name or
  contents
• Windows encodes type in name
• .com, .exe, .bat, .dll, .jpg, etc.
• Unix encodes type in contents
• Magic numbers, initial characters (e.g., ! for
  shell scripts)

7
File system structure.

• To improve IO efficiency, transfer of data is
  made in units of blocks.
• Each block is one
• or more disk sectors.
• Blocks vary in size.

8
File system structure.

• Disk have two main characteristic making them
  convenient storing devices
• Can be rewritten in place.
• Read from disk, modify, write back to same disk
  location.
• Random access.
• Any block can be accessed directly, consequently
  any file can be accessed.
• Once accessed, following accesses can be
  performed sequentially or randomly.
• Move to another file is just a matter of moving
  disk read-write head to proper location.

9
A Typical File-system Organization
10
File system organization.

• FS is a logical layer over the disk structure.
• Allow easy operations of store, locate and
  retrieve.
• File system organization involves
• Definition of file its attributes and permissions
• Operations allowed on files
• Directory structure for file organization.
• Algorithms and data structures to map logical
  file system on top of the physical device.

11
File system organization.

• File system is consist of some levels

12
File system organization.

• Logical FS uses directory structure to provide
  file-organization module with needed info.
• It is also responsible for protection and
  security issues.
• File organization module translate logical block
  address to physical block address.
• This module also include free space manager.
• Basic FS issue generic commands to read and write
  physical blocks on disk.
• For example drive1, cylinder x, track y sector z

13
File system layout

• How do file systems use the disk to store files?
• File systems define a block size (e.g., 4KB)
• Disk space is allocated in granularity of blocks
• A Master Block determines location of root
  directory
• Always at a well-known disk location
• Often replicated across disk for reliability
• A free map determines which blocks are free,
  allocated
• Also stored on disk, cached in memory for
  performance
• Remaining disk blocks used to store files (and
  dirs)
• There are many ways to do this

14
File system interface

• File data access
• READ Bring a specified chunk of data from file
  into the process virtual address space
• WRITE Write a specified chunk of data from the
  process virtual address space to the file
• CREATE, DELETE, SEEK, TRUNCATE
• open, close, set_attributes
• Many semantical issues
• Automatic size-extension
• Holes
• Persistence of open files
• More

15
Allocation methods

• One of file handling problems is how to allocate
  disk space.
• Contiguous allocation
• File should reside in a contiguous disk blocks.
• Efficient for reading files.
• Many problems
• File size might be unknown at creation, or grow
  over time.
• Dynamic handling of disk space (first-fit,
  best-fit)
• External fragmentation

16
Allocation methods

• Linked allocation
• Each file is a linked list of disk blocks.
• Directory contain pointer to first and last file
  block.
• File can be scattered all over disk (performance)
• direct acces to middle of file is problematic.
• Creation of new file is allocation directory
  entry with null pointer to first block, as file
  grow blocks are added to list.
• No external fragmentation.
• Reliability if pointer in middle of file is
  lost, rest of file is lost.

17
Allocation methods linked allocation
18
Allocation methods FAT fs

• Variant of linked list is FAT fs
• FAT file allocation table.
• A section at the part of each partition is
  dedicated to the table.
• Table has one entry for each disk block and is
  indexed by block no.
• Directory contain first block no
• Rest are retrieved from table.
• Unused blocks are indexes with zero value
  (maintained in linked list for easy search)
• Unless table is cached, many head seeks might
  occur.
• Find of next block in table, accessing block.

19
Allocation methods FAT fs
20
Allocation methods Indexed allocation

• Indexed allocation
• Linked allocation support for direct access is
  poor.
• Indexed allocation solve this problem by grouping
  all block pointers together in one location
• index block also called I-node.
• Directory contain location of index block.
• Direct access is achieved from index block.
• No external fragmentation.
• Suffer from wasted space
• Index block need much more disk space than
  pointers of linked list.

21
Allocation methods Indexed allocation
22
Allocation methods Indexed allocation

• Unix I-node implementation
• Direct access pointers
• Several levels of indirection tables to blocks.
• Allow one file to exceed size of 232 bytes.

23
I-node
24
Free space management

• Bit vector
• Hold bit vector for all blocks
• Each bit indicate if corresponding block is
  free/allocated.
• Linked list.
• All free blocks are placed in one linked list.
• Grouping
• Store address of n free blocks in 1st free block.
• Block n-1 is the address where next n addresses
  are stored and so on.

25
Free space management

• Counting
• Take advantage of the fact that usually some
  contiguous blocks are allocated/free together.
• Keep address of first block and number of free
  blocks.
• Each such cluster is a linked list node.

26
Directories

• Directories serve two purposes
• For users, they provide a structured way to
  organize files
• For the file system, they provide a convenient
  naming interface that allows the implementation
  to separate logical fil organization from
  physical file placement on the disk
• Most file systems support multi-level directories
• Naming hierarchies (/, /usr, /usr/local/, )
• Most file systems support the notion of a current
  directory
• Relative names specified with respect to current
  directory
• Absolute names start from the root of directory
  tree

27
Operations Performed on Directory

• Search for a file
• Create a file
• Delete a file
• List a directory
• Rename a file
• Traverse the file system

28
Organize the Directory (Logically)

• Efficiency locating a file quickly
• Naming convenient to users
• Two users can have same name for different files
• The same file can have several different names
• Grouping logical grouping of files by
  properties, (e.g., all Java programs, all games,
  )

29
Directory organization

• A single directory for all users

Naming problem Grouping problem
30
Directory organization

• Two level directories
• Separate directory for each user

Path name Can have the same file name for
different user Efficient searching
31
Directory organization

• Tree structured directory.

32
Directory organization

• A-cyclic graph directories
• Can share files and directories.

33
Directories structure in Unix

• Full path is not attribute of file
• Same file can be reached from different semantic
  paths.
• Use-counter is maintained in order to identify
  when file need to be actually deleted.
• Process has data structure of open files.

34
File System Mounting

• Same as file need to be opened for use, A file
  system must be mounted before it can be accessed.
• A un-mounted file system is mounted at a mount
  point
• OS is given name of device and location with in
  directory to mount it.

35
File System Mounting
(a) Existing (b)unmounted partition
36
File System Mounting
Mount point
37
Protection

• File systems implement some kind of protection
  system
• Who can access a file
• How they can access it
• More generally
• Objects are what, subjects are who, actions
  are how
• A protection system dictates whether a given
  action performed by a given subject on a given
  object should be allowed
• You can read and/or write your files, but others
  cannot
• You can read /etc/motd, but you cannot write it

38
Representing protection

• Access Control Lists (ACL)
• For each object, maintain a list of subjects and
  their permitted actions

• Capabilities
• For each subject, maintain a list of objects and
  their permitted actions

39
Files Access control Lists and Groups haring

• Mode of access read, write, execute
• Three classes of users
• RWX
• a) owner access 7 ? 1 1 1 RWX
• b) group access 6 ? 1 1 0
• RWX
• c) public access 1 ? 0 0 1
• Ask manager to create a group (unique name), say
  G, and add some users to the group.
• For a particular file (say game) or subdirectory,
  define an appropriate access.

Attach a group to a file chgrp G
game
40
Recovery

• Consistency check
• Part of memory is cached to speed up access.
• Directory info in cache is more up to date then
  on disk
• Cache was not updated yet on disk
• Computer crash might result in lost of valuable
  directory info.
• File system might be left inconsistent
• Usually a disk consistency check is run at boot
  time
• Comparing directory data with block data on disk
• Try to fix problems
• If directory entry is lost in I-node system it
  might result in lost of file.
• Data blocks have no knowledge on one another
• Thats why UNIX cache for read but force write to
  disk on update.

41
Recovery

• Use system programs to back up data from disk to
  another storage device (floppy disk, magnetic
  tape).
• Recover lost file or disk by restoring data from
  backup.

42
Recovery by log

• 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.
• The transactions in the log are asynchronously
  written to the file system. When the file system
  is modified, the transaction is removed from the
  log.
• If the file system crashes, all remaining
  transactions in the log must still be performed.

43
Recovery by log

• pros.
• Asynchronous write to disk.
• Efficient recovery
• Depends on number of transactions in log file and
  not size of FS.
• Cons.
• Doubles number of writes to disk.