Access Methods File store information

When it is used it is accessed & read into memory

Some systems provide only one access method

IBM support many access methods

Choosing right one for a particular application

Access Methods Sequential Access

read nextwrite next resetno read after last write

(rewrite) Direct Access

read nwrite nposition to n

read nextwrite next

rewrite nn = relative block number

Sequential-access FileSimplest access method

Information accessed- one record after the otherCommon method – ex: editors & compilers

Bulk of operations in file is read & writeRead – reads the next portion of the file & advances file pointerWrite – appends the end of the file & advances to the newly written material

Beginning – resetForward & backward n records – skip

Sequential access is based on the tape model of a file

Sequential-access File

Simulation of Sequential Access on a Direct-access File

Direct Access (relative access)file – fixed length, logical records (allow programs to read & write

rapidly) , no particular orderRandom access to any file blockBased on disk model of a fileFile is viewed as a numbered sequence of blocks / recordsNo restrictions – read 24, read 54, write 5Block number provided by user to OS is normally a relative block

numberIt is an index relative to index of a fileRelative blocks – 0,1,… / 1,………..Use – allows OS to decide where the file should be placed &helps

users from accessing file part that is not a part of itLogical record length – L request for record – N turns to I/O request for L bytes starting from location L*(N-1) within the fileLogical records are of fixed size – easy to read, write & delete a record

Other access methodsThey can built on the top of the direct access methodInvolve the construction of index for a fileContains pointers for various blocksFind a record in a file – search the index – use the pointer to access the

file directly – find the desired recordEx: retail price fileLarge files – index file will be large to be to kept in memorySolution – create an index for the indexed filePrimary index file contain pointers to secondary index file which would point actual data itemEx: IBM’s indexed sequential access method (ISAM) uses small master

index for secondary indexFind a item: binary search in master index for finding the block number

in secondary index – block is read – again binary search to find the block containing the desired record – block is searched

sequentiallyAny records can be accessed by this two direct access reads

Example of Index and Relative Files

Directory StructureA collection of nodes containing information about all files

F 1 F 2F 3

F 4

F n

Directory

Files

Both the directory structure and the files reside on diskBackups of these two structures are kept on tapes

The file systems of computer can be extensiveStore – millions of files (TB), manage – organizeOrganization – 2 parts:

Disk split into 1/more partitions – minidisksPartition – low level structure – files & directories

residePartition contains information about files in itInformation is kept in device directory / volume table

of contentsDevice dir – records name , location, size, typeDirectory – symbol table – translates file name into

directoriesInsert, delete, search, list all entries

A Typical File-system Organization

Operations Performed on DirectorySearch for a file – dir struct – entry for a fileCreate a file – new file created & added to

dirDelete a file –no longer needed remove

from directoryList a directory- list all the filesRename a file – changing the nameTraverse the file system – checking the

entire directory

Organize the Directory (Logically) to Obtain

Efficiency – locating a file quicklyNaming – 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, …)

Single-Level DirectoryA single directory for all users

Naming problem

Grouping problem

Two-Level DirectorySeparate directory for each user

Path name Can have the same file name for different user Efficient searching No grouping capability

Tree-Structured Directories

Tree-Structured Directories (Cont)Efficient searching

Grouping Capability

Current directory (working directory)cd /spell/mail/progtype list

Tree-Structured Directories (Cont)

Absolute or relative path nameCreating a new file is done in current directoryDelete a file

rm <file-name>Creating a new subdirectory is done in current

directorymkdir <dir-name>

Example: if in current directory /mailmkdir count

mail

prog copy prt exp count

Deleting “mail” deleting the entire subtree rooted by “mail”

Acyclic-Graph DirectoriesHave shared subdirectories and files

Acyclic-Graph Directories (Cont.)Two different names (aliasing)

If dict deletes list dangling pointerSolutions:Backpointers, so we can delete all pointers

Variable size records a problemBackpointers using a daisy chain organizationEntry-hold-count solution

New directory entry typeLink – another name (pointer) to an existing fileResolve the link – follow pointer to locate the file

General Graph Directory

General Graph Directory (Cont.)How do we guarantee no cycles?

Allow only links to file not subdirectoriesGarbage collectionEvery time a new link is added use a cycle

detectionalgorithm to determine whether it is OK