File Operations

O File is an abstract data type

O Create
O Write – at write pointer location
O Read – at read pointer
location O Reposition within file -
seek
O Delete
O Truncate
O Open(Fi) – search the directory structure on
disk for entry Fi, and move the content of
entry to memory
O Close (Fi) – move the content of entry Fi in
memory to directory structure on disk
Open Files
O Several pieces of data are needed
to manage open files:
O Open-file table: tracks open files
O File pointer: pointer to last read/write
location, per process that has the file open
O 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
O Disk location of the file: cache of data
access information
O Access rights: per-process access mode
information
Open File Locking
O Provided by some operating systems and
file systems
O Similar to reader-writer locks
O Shared lock similar to reader lock –
several processes can acquire
concurrently
O Exclusive lock similar to writer lock
O Mediates access to a file
O Mandatory or advisory:
O Mandatory – access is denied depending on
locks held and requested
O Advisory – processes can find status of locks
and decide what to do
File Types – Name, Extension
Directory Structure
O A collection of nodes
containing information about all
files
Disk Structure
O Disk can be subdivided into partitions
O Disks or partitions can be RAID protected
against failure
O Disk or partition can be used raw – without a
file system, or formatted with a file system
O Partitions also known as minidisks, slices
O Entity containing file system known as a
volume
O Each volume containing file system also tracks that
file system’s info in device directory or volume
table of contents
O As well as general-purpose file systems
there are many special-purpose file
systems, frequently all within the same
operating system or computer
A Typical File-system
Organization
Types of File Systems
O We mostly talk of general-purpose file systems
O But systems frequently have may file systems,
some general- and some special- purpose
O Consider Solaris has
O tmpfs – memory-based volatile FS for fast,
temporary I/O
O objfs – interface into kernel memory to get
kernel symbols for debugging
O ctfs – contract file system for managing
daemons
O lofs – loopback file system allows one FS to be
accessed in place of another
O procfs – kernel interface to process structures
O ufs, zfs – general purpose file systems
Operations Performed on
Directory
O Search for a file

O Create a file

O Delete a file

O List a directory

O Rename a file

O Traverse the file

system
 Directory Organization
The directory is organized logically to obtain

O Efficiency – locating a file quickly

O Naming – convenient to users
O Two users can have same name for
different files
O The same file can have several different
names
OGrouping – logical grouping of files by
properties, (e.g., all Java programs, all games,
…)
Single-Level Directory
O A single directory for all
users

O Naming problem
O Grouping problem
Two-Level Directory
O Separate 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.)
O Efficient searching

O Grouping Capability

O Current directory (working directory)

O cd /spell/mail/prog
O type list
 Tree-Structured Directories
(Cont)
O Absolute or relative path name
O Creating a new file is done in current directory
O Delete a file
rm <file-name>
O Creating a new subdirectory is done in
current directory
mkdir <dir-name>
Example: if in current directory
/mail mkdir count

Deleting “mail”  deleting the entire

subtree rooted by “mail”
Acyclic-Graph Directories
O Have shared subdirectories and
files
Acyclic-Graph Directories
(Cont.)
O Two different names (aliasing)
O If dict deletes list  dangling
pointer
O Solutions:
O Backpointers, so we can delete all
pointers Variable size records a problem
O Backpointers using a daisy chain
organization
O Entry-hold-count solution
O New directory entry type
O Link – another name (pointer) to an existing
file
O Resolve the link – follow pointer to locate
the file
General Graph Directory
General Graph
Directory (Cont.)
O How do we guarantee no cycles?
O Allow only links to file not subdirectories
O Garbage collection
O Every time a new link is added use a cycle
detection algorithm to determine whether
it is OK
File System Mounting
O A file system must be mounted before it can
be accessed
O A unmounted file system (i.e., Fig. 11-11(b)) is
mounted at a mount point
File Sharing
O Sharing of files on multi-user systems is desirable
O Sharing may be done through a protection
scheme
O On distributed systems, files may be shared across
a network
O Network File System (NFS) is a common
distributed file-sharing method
O If multi-user system
O User IDs identify users, allowing permissions
and protections to be per-user
Group IDs allow users to be in groups, permitting group
access rights
O Owner of a file / directory
O Group of a file / directory
File Sharing – Remote
File Systems
O Uses networking to allow file system access between systems
O Manually via programs like FTP
O Automatically, seamlessly using distributed file systems
O Semi automatically via the world wide web
O Client-server model allows clients to mount remote file systems
from servers
O Server can serve multiple clients
O Client and user-on-client identification is insecure or
complicated
O NFS is standard UNIX client-server file sharing protocol
O CIFS is standard Windows protocol
O Standard operating system file calls are translated into remote
calls
O Distributed Information Systems (distributed naming services) such
as LDAP, DNS, NIS, Active Directory implement unified access to
information needed for remote computing
File Sharing – Failure Modes
O All file systems have failure modes
O For example corruption of directory structures
or other non-user data, called metadata
O Remote file systems add new failure
modes, due to network failure, server
failure
O Recovery from failure can involve state
information about status of each
remote request
O Stateless protocols such as NFS v3 include
all information in each request, allowing
easy recovery but less security
File Sharing – Consistency
Semantics
O Specify how multiple users are to access a shared
file simultaneously
O Similar to Ch 5 process synchronization algorithms
O Tend to be less complex due to disk I/O and network
latency (for remote file systems
O Andrew File System (AFS) implemented complex
remote file sharing semantics
O Unix file system (UFS) implements:
O Writes to an open file visible immediately to other users
of the same open file
O Sharing file pointer to allow multiple users to read
and write concurrently
O AFS has session semantics
O Writes only visible to sessions starting after the file is
closed
Protection
O File owner/creator should be able to
control:
O what can be done
O by whom
O Types of access
O Read
O Write
O Execute
O Append
O Delete
O List
Access Lists and
Groups
O
O
Mode of access: read, write, execute
Three classes of users on Unix / Linux
RWX
a) owner access 7  1
11 RWX

b) group access 6  1 RWX

10
O Ask manager to create a group (unique name), say G, and add
c) public access
some users 1
to the group.  0
01
O For a particular file (say game) or subdirectory, define an
appropriate access.

Attach a group to a file chgrp G game

Windows 10 Access-
Control List Management
A Sample UNIX Directory
Listing
Solid-State Drive (SSD)
O Nonvolatile memory
O This gives the device the “solid-state” name
O Most common media: NAND Flash memory (Floating-
gate transistors)
O
SSD: 2013 - 2015
O Intel 520 Cherryville (2013) O Samsung 850
O Capacity: 240 GB (2015)
O Compare to HD: TB
O Capacity: 1TB
O Sequential R/W: 550/520
MB/s O Sequential R/W: 540/520
O Compare to HD: 122 MB/s
MB/s
O Latency
O Latency
O Read: 0.04 ms O N/A
O Write: 0.2 ms O $0.5/GB ($0.2/GB
O Compare to HD: tens today)
of ms
O $1/GB
O Compare to HD: 0.06/GB
Limitations of SSD
O Random write performance
O For write, need to first erase a large of
pages (32-64 pages), and reprogram
O Burnout
O Each cell has limited erase/program
cycles
O Range from 1,000 – 100,000 writes

© 2016-2017 DING YUAN ALL RIGHTS RESERVED

How do SSD’s characteristics
impact FS design?
O Characteristics of SSD
O No mechanical components ---
data location won’t make
difference
• Optimizations on data location are no
longer useful

O Random write performance is BAD

• Avoid random writes!
O Limited “write” cycles for each cell
• Spread the writes across the entire
SSD
© 2016-2017 DING YUAN ALL RIGHTS RESERVED