Scribd
Upload
33 views

Storage 4

The document discusses file systems and how they organize files on disk. It describes different methods for allocating file blocks, including contiguous, linked list, and indexed allocation. Contiguous allocation stores file blocks together but cannot easily grow files. Linked list allocation allows growing files but has poor random access. Indexed allocation uses an index block to point to other blocks, enabling fast random access but making large growth clumsy.

Uploaded by

Harshit Garg
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
33 views

Storage 4

The document discusses file systems and how they organize files on disk. It describes different methods for allocating file blocks, including contiguous, linked list, and indexed allocation. Contiguous allocation stores file blocks together but cannot easily grow files. Linked list allocation allows growing files but has poor random access. Indexed allocation uses an index block to point to other blocks, enabling fast random access but making large growth clumsy.

Uploaded by

Harshit Garg
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 15

File System

File
 File:
 Contiguous logical address space to store data / program.
 Contiguous / Non Contiguous blocks on HDD.
 File Attributes: (Metadata of file)
 Name
 Inode Number
 Type
 Location
 Size
 Protection
 Time, date, and user identification
File Operations
 Open
 Create
 Close

#include <fcntl.h>
#include <unistd.h>
#include <sys/types.h>

int open (const char *filename, int flags [, mode_t mode])


int create (const char *filename, mode_t mode)
int close (int filedes)

Bit vector of: Bit vector of Permission Bits:


• Access modes (Rd, Wr, …) • User |Group|Other X R|W|X
• Open Flags (Create, …)
• Operating modes (Appends,
…)
File Operations
 Read
 Write
 Seek
 Delete

ssize_t read (int filedes, void *buffer, size_t maxsize)


- returns bytes read, 0 => EOF, -1 => error

ssize_t write (int filedes, const void *buffer, size_t size)


- returns bytes written

off_t lseek (int filedes, off_t offset, int whence)


File System
 File System is method for storing and organizing computer files such that its
easy to find and access them later.
 File System Components
 Disk Management:
 Mapping files to blocks and vice versa
 Keep track of free blocks
 Naming:
 Map file name to its blocks
 Protection:
 Layer to keep data secure
 Reliability/Durability:
 Keeping of files durable despite crashes, media failures, attacks, etc
Translating from User to System View

File
System

 User says: “Give me bytes 2—12”?


Fetch block corresponding to those bytes
Return just the correct portion of the block
 What about: “write bytes 2—12”?
 Fetch block
 Modify portion
 Write out Block
 Everything inside File System is in whole size blocks
 For example, getc(), putc()  buffers something like 4096 bytes,
even if interface is one byte at a time
 From now on, file is a collection of blocks
Disk Management Policies
 What we store on the disk:
 File
 user-visible group of blocks arranged sequentially in logical space
 Directory
 user-visible index mapping names to files
 Actually a file only
 Need way to structure files: File Header / Inode
 Track which blocks belong at which offsets within the logical file
structure
 Optimize placement of files disk blocks to match access and usage
patterns
Disk Management Policies
 Need way to track free disk blocks
 Link free blocks together
 Maintain a linked list
 Slow
 Use bitmap to represent free space on disk
 Will consume memory
Disk Management Policies
 Access disk as linear array of sectors:
 Identify sectors as vectors [cylinder, surface, sector].
 Sort in cylinder-major order.
 OS must deal with bad blocks
 Not used much anymore.
 Logical Block Addressing (LBA).
 Every sector has integer address from zero up to max number of
sectors.
 Controller translates from LBA to physical position
 OS need not worry about disk structure
File System Access Patterns
 How do user access file?
 Need to know the access pattern user is likely to throw at system.
 Sequential Access:
 Bytes read in ordered fashion.
 “Give me “X” bytes”, then give me next “Y” bytes …
 Almost all file access are of this type.
 Random Access:
 Read / write from the middle of file.
 “Give me bytes “I” to “J”
 Don’t have to read the complete file for this.
 Less frequent, but still important.
 Content based Access:
 “Find 100 bytes starting with BITS”
 Structured data can provide this functionality.
 Build indexes on the data
File System based on Usage Patterns
 Facts:
 Most files are small.
 Very few files are big.
 Large files use up most of the disk space and bandwidth.
 Few enormous files are equivalent to an immense number of small files.
 File Size Distribution on UNIX Systems—Then and Now
 Andrew S. Tanenbaum, Jorrit N. Herder, Herbert Bos
 A Large-Scale Study of File-System Contents
 John R. Douceur and William J. Bolosky
How to organize files on disk
 Goals:
 Maximize Sequential Performance
 Effective Random Access Performance
 Easy Management of file (grow, truncate etc)
Block Allocation Methods - Contiguous
 Allocate continuous range of blocks on HDD
 User pre-declares max file size (disadvantage)
 Search bit-map for space using best fit/first fit
 What if not enough contiguous space for new file?
 File Header Contains:
 First sector/LBA in file
 File size (# of sectors)
 Pros:
 Fast Sequential Access
 Easy Random access
 Cons:
 External Fragmentation
 Hard to grow files
 Could compact space, but that would be really expensive
 Continuous Allocation was used by IBM 360
Linked List Allocation
 Each block points to next block on the disk.

File Header

 Pros:
Null
 Can grow files dynamically
 Free list same as file
 Cons:
 Bad Sequential Access (seek between each block),
 Unreliable (lose a block -> lose rest of file)
 Serious Con: Bad random access!!!!
 Technique originally from Alto (First PC, built at Xerox)
 No attempt to allocate contiguous blocks
Indexed Allocation

 System allocates file header block to hold array of pointers big enough to
point to all data blocks
 User pre-declares max file size
 Pros:
 Can easily grow up to space allocated for index
 Random access is fast
 Cons:
 Clumsy to grow file bigger than table size
 Still lots of seeks, blocks may be spread over disk

You might also like