Lesson 7 (File System Structure)

Created by

qothrunnada

Cards (39)

Fille Attributes
Name – only information kept in human-readable form
Identifier – unique tag (number) identifies file within file system
Type – needed for systems that support different types
Location – pointer to file location on device
Size – current file size
Protection – controls who can do reading, writing, executing
Time, date, and user identification – data for protection, security, and usage monitoring
Information about files are kept in the directory structure, which is maintained on the disk
Many variations, including extended file attributes (e.g. checksum)
File Operations
Create
Write – at write pointer location
Read – at read pointer location
Reposition within file – seek (move the file pointer)
Delete – release all file space used by the file and erase the directory entry
Truncate – maintain the file attribute but erase the contents in the file
Open(F) – search the directory structure on disk for entry F, and move the content of entry to memory
Close (F) – move the content of entry F in memory to directory structure on disk
who "Opens" a file and when does it occur?
Before reading/writing a file, kernel must "open" the file.
What does "open" file operation do?
locate the file on the disk according to its file name (path);
establish data structures to share file among different processes
allocate file descriptor(ID-used by read/write requests)
What are the data needed to manage open files?
Open-file table: tracks open files
File pointer: pointer to last read/write location, per process that has the file open
File-open count: counter of number of times a file is open – to allow removal of data from open-file table when last processes closes it
Disk location of the file: cache of data access information
Access rights: per-process access mode information
What are the 3 data structures in a UNIX system?
Per Process Table of Open Files
System-wide File Table
System-wide V-Node Table (Linux –no V-node, only I-node)
What is inside a Per Process Table of Open Files?
Each entry for an opened file
• File descriptor Flag (fd flag): indicates whether the file should be closed upon executing one of the “exec” functions
Ptr: point to entry in the System Wide File Table(describes how to access the file)
What happens in Per Process Table when a new process is created?
3 std file is opened
Standard input(stdin, fd=0) eg. keyboard
Standard output(stdout, fd=1) eg. display
Standard error(stderr, fd=2)
What is a System Wide Open File TableTable for all opened files for all processes
What is a System Wide Open File Table?
A table for all opened files for all processes
What is inside a System Wide Open File Table
File Status Flags (file open for):
read only (O_RDONLY)
write only (O_WRONLY)
both read and write (O_RDWR) + (Optional)
append on each write (O_APPEND)
File offset(position of next read/write)
V-node ptr (entry in V-node table) or I-node ptr (entry in I-node table)
V-node Table
Each opened file = only 1 entry in the table
V-node info: • Type of file (regular, pipe, directory, etc) • Ptr to function operates such type of file
I-node info: • Only for regular & directory file • Owner, access permission, size, access time • Ptr to the actual data blocks on disk for the file
Types of Sharing of Files
Type 1: File A is shared via 2 separate file table entries
Type 2: File B is shared via the same file table entry
Type 1 Sharing between 2 process
File is shared via 2 separate file table entires
Each process can have a different file status and offset for the same file (e.g read and write)
Type 2 Sharing between 2 processes
File is shared via the same file table entry
Created via dup or fork where >1 fd point to the the same file table entry
Share the same file status (read, write, RW) and offset
What does each file's i-node contain?
file type
access permissions
Owner and group owner
Size of the file
number of links
device number of the device where the file is stored
time of last access
time of last modification
time of last change to the i-node
block numbers of the blocks in which file is stored
how to see file i-node number?
ls -i <file>
stat <file>
How are file identified in a Unix Kernel?
By its i-node
What does a UNIIX directory file contain and what does it do?
Unix directory file contains: names of the files, their i-node numbers and contents of the directory file
Provides a mapping between file name & its i-node number:
contents of the directory file (“assign1”) would be something like
What is (Open) File Locking
A mechanism that allows one to lock a file and prevent other processes from gaining access to it (provide by some OS and file system)
Types of File Locking
Shared Lock: similar to reader lock, Where several processes/user can can access a file simultaneously for reading purposes only.
Exclusive Lock: similar to writer lock, Only one process/user can access a file simultaneously for reading purposes only.
Mandatory Locks: Access is denied by OS depending on locks held and requested.
Advisory – Processes can find status of locks and decide what to do (OS will not enforce but decide by the programmer, programmer can ignore)
File Type (Name and Extension)
FYI
Type of File Structure
None (no structure) - a sequence of words/bytes
Structure:
Simple Record Structure (Lines, Fixed Length, Variable Length)
Complex Structure (Formatted Doc, Relocatable Load File - e.g executable file)
Why do we need structured?
The OS needs to determine where in the RAM to load the executable file and where is the location of first instruction
Types of File Access Methods
Sequential
Direct
Indexed
(Type of Access Method) - Sequential
Information in the file is processed in order, one record after the other
Related variables: read_next, write_next, reset
Type of Access Methods (Others)
Can be built on top of base methods
Example: creation of an index for the file which contains pointer to the actual disk block • Keep index in memory for fast determination of location of data to be operated on • If index file is too large, create index (in memory) of the index (on disk)
Example: IBM indexed sequential-access method (ISAM), VMS OS
DIrectory Structure
What is a Directory? A collection of nodes containing information about all files
Where do they both reside? On the disk
Operations Performed on Directory
Search for a file (within the directory)
Create a file – nano, vi
Delete a file - rm
List a directory - ls
Rename a file - mv
Traverse the file system - cd
How is a Direrectory Organised?
Efficiency – locating a file quickly
Naming – convenient to users
Two users can have same name for different files
user1 in his own directory create a file “xyz.txt”
user 2 in his own directory create another file also named it as “xyz.txt”
The same file can have several different names
Grouping – logical grouping of files by properties, (e.g., all Java programs, all games, …)
Types of Directories
Single - Level
Two - Level
Tree - Structured
Acyclic - Graph
General Graph
(Types of Directory) Single-Level Directory
A single directory for all users
ALL files are kept in the same directory
Benefits:
simplest directory structure
actions like file creation, searching, file deletion, and updating are quite simple
Problems:
Naming - need to have unique name for all files
Grouping - can’t differentiate between groups
(Types of Directory) - Two-Level Directory
Separate directory for each user
Features:
Path name (i.e user name + file name)
Can have same file name for different user
Efficient searching
Disadvantage:
No grouping capability: can't create group (sub-folder) under the same user
Cannot share: user cannot share a file with another user without permission.
Not scalable: files of the same type cannot be kept in the same group
(Types of Directory) - Tree-Structured Directories (multi-level)
Features/Benefits:
Efficient Searching: Uses absolute or relative path name
Grouping Capabilities: users can define own subdirectory for files of similar type
delete/create file/directory is done in current one. deleted a directory deletes the entire subtree rooted by specified directory
Disadvantage:
Inefficient: Files may be saved in numerous directories if they do not fit into the hierarchical structure.
Cannot share files between users.
(Types of Directory) - Acyclic-Graph Directory
Features:
2 or more directory entries can lead to the same subdirectory / file and prevents any entries from pointing back to directory structure.
Structure allows for file sharing
Uses links:
symbolic / soft link (specify a file path: logical)
hard link (actual link to the same file on the disc from multiple directories: physical).
Advantages: Due to different-different paths, searching is simple
Disadvantages (when removing):
Softlink: a dangling pointer remains.
Hardlink: must erase all the references associated with it.
(Type of Directory) General Graph DIrectory
Features:
Cycles are allowed inside directory structure where numerous directories can be derived from more than one parent directory.
Commands like, "search" must be used with caution. If cycles are allowed, the search is infinite.
Advantages: More adaptable than the others
Disadvantages:
It is more expensive than other options.
Garbage collection is required.
Hard to figure out how much space the files and folders have used up
(Type of Directory) - General Graph Cycle (2) 
How do we guarantee no cycles?
Allow only links to file not subdirectories
Garbage collection (traversing the entire file system to locate no accessible entry)
Every time a new link is added use a cycle detection algorithm to determine whether it is OK
File System Mounting
A file system (eg. USB) must be mounted before it can be accessed.
A unmounted file system is mounted at a mount point.
File Protection (Creation and Access Types)
File owner/creator should be able to control: what can be done and by whom
Types of access:
Read
Write
Execute
Append
Delete
List
File Protection (Access Lists and Groups)
• Mode of access: read, write, execute
Three classes of users on Unix / Linux: