OPERATINGSYSTEMS

LECTURENOTES

(2023-2024)

DEPARTMENT OF
COMPUTERSCIENCEANDENGINEERING
(R18A0504)OPERATINGSYSTEMS
OBJECTIVES:
 TolearnthefundamentalsofOperatingSystems.
 TolearnthemechanismsofOStohandleprocessesandthreadsandtheir communication
 TolearnthemechanismsinvolvedinmemorymanagementincontemporaryOS
 Togainknowledgeondistributedoperating systemconceptsthat includes architecture,
 Mutualexclusionalgorithms, deadlockdetectionalgorithmsandagreementprotocols
 Toknowthecomponentsandmanagementaspectsofconcurrencymanagement

UNIT-I
Introduction: Concept of Operating Systems, Generations of Operating systems, Types of
Operating Systems, OS Services, System Calls, Structure of an OS - Layered, Monolithic,
Microkernel Operating Systems, Concept of Virtual Machine. Case study on UNIX and
WINDOWS Operating System.
Processes: Definition, Process Relationship, Different states of a Process, Process State
transitions, Process Control Block (PCB), Context switching
Thread: Definition, Various states, Benefits of threads, Types of threads, Concept of
Multithreads.

UNIT-II
Process Scheduling: Foundation and Scheduling objectives, Types of Schedulers, Scheduling
criteria: CPU utilization, Throughput, Turnaround Time, Waiting Time, Response Time;
Scheduling algorithms: Pre-emptive and Non pre-emptive, FCFS, SJF, RR; Multiprocessor
scheduling: Real Time scheduling: RM and EDF.
Inter-process Communication: Critical Section, Race Conditions, Mutual Exclusion,Hardware
Solution, Strict Alternation, Peterson’sSolution, TheProducer/Consumer Problem, Semaphores,
Event Counters, Monitors, Message Passing, Classical IPC Problems: Reader’s & Writer
Problem, Dinning Philosopher Problem etc.

UNIT-III
Memory Management: Basic concept, Logical and Physical address map, Memory allocation:
Contiguous Memory allocation – Fixed and variable partition–Internal and External
fragmentation and Compaction; Paging: Principle of operation – Page allocation – Hardware
support for paging, protection and sharing, Disadvantages of paging.
Virtual Memory: Basics of Virtual Memory – Hardware and control structures – Locality of
reference, Page fault , Working Set , Dirty page/Dirty bit – Demand paging, Page Replacement
algorithms: Optimal, First in First Out (FIFO), Second Chance (SC), Not recently used (NRU)
and Least Recently used (LRU).
UNIT-IV
File Management: Concept of File, Access methods, File types, File operation, Directory
structure, File System structure, Allocation methods (contiguous, linked, indexed), Free-space
management (bit vector, linked list, grouping), directory implementation (linear list, hash table),
efficiency and performance.
I/OHardware:I/Odevices,Devicecontrollers,Direct memoryaccessPrinciplesofI/O
Software:GoalsofInterrupt handlers,Devicedrivers,Device independent I/Osoftware.

UNIT-V
Deadlocks: Definition, Necessary and sufficient conditions for Deadlock, Deadlock Prevention,
Deadlock Avoidance: Banker’s algorithm, Deadlock detection and Recovery.
DiskManagement:Diskstructure,Diskscheduling-FCFS,SSTF,SCAN,C-SCAN,Disk reliability, Disk
formatting, Boot-block, Bad blocks.

TEXTBOOKS:
1. OperatingSystemConceptsEssentials,9thEditionbyAviSilberschatz,Peter Galvin,Greg
Gagne, Wiley Asia Student Edition.
2. Operating Systems: Internals and Design Principles, 5th Edition, William
Stallings,Prentice Hall of India.

REFERENCEBOOKS:
1. OperatingSystem:ADesign-orientedApproach,1stEditionbyCharlesCrowley, Irwin
Publishing
2. OperatingSystems:AModernPerspective,2ndEditionbyGaryJ.Nutt,Addison- Wesley
3. Design of the Unix Operating Systems, 8th Edition by Maurice Bach, Prentice-Hallof
India
4. Understanding the Linux Kernel, 3rd Edition, Daniel P. Bovet, Marco Cesati, O'Reilly
and Associates

OUTCOMES:
At theend of the coursethestudentsareable to:
 Createprocessesand threads.
 Develop algorithms for process scheduling for a given specification of CPUutilization,
Throughput, Turnaround Time, Waiting Time, Response Time.
 For agivenspecificationofmemoryorganizationdevelop thetechniques foroptimally
allocating memory to processes by increasing memory utilization and for improving
the access time.
 Designandimplement filemanagement system.
 ForagivenI/OdevicesandOS(specify)developtheI/O management functions inOS as
part of a uniform device abstraction by performing operations for synchronization
between CPU and I/O controllers.
 INDEX
UNIT
TOPIC PAGENO
NO
Introduction

Operating Systemconcepts 1-11

TypesofOperatingSystems 11-18
I
Operatingservices,SystemCalls 18-25

StructureofOS,Virtual machines 26-31

ProcessConcepts 32-34
Thread Concepts 34-38
ProcessScheduling

ProcessScheduling concepts 39-40

Pre-emptiveandNonpre-emptivescheduling
41-48
algorithms
II Multiprocessorscheduling 48-49
Realtimescheduling 49-52
Inter-processCommunication
CriticalSection problem 52-57
ClassicalIPC Problems 57-65
MemoryManagement 66-82
III
VirtualMemory 82-89
FileSystem Management 90-105
IV
I/OHardware 105-110
Deadlocks 111-119
V
MassStorageStructure 120-129
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

UNIT-I
Operating System Introduction: Operating Systems Objectives and functions, Computer System
Architecture, OS Structure, OS Operations, Evolution of Operating Systems - Simple Batch, Multi
programmed, time shared, PersonalComputer, Parallel, Distributed Systems, Real-Time Systems, Special -
Purpose Systems, Operating System services, user OS Interface, System Calls, Types of System Calls,
System Programs, Operating System Design and Implementation, OS Structure, Virtual machines

1
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

2
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Operatingsystemperformsthefollowing functions:

1. Booting
Booting isaprocessofstartingthecomputer operatingsystemstartsthecomputerto work. It
checks the computer and makes it ready to work.
2. Memory Management
It is also an important function of operating system. The memory cannot be managed
without operating system. Different programs and data execute in memoryat one time. if
there is no operating system, the programs may mix with each other. The systemwill not
work properly.
3. LoadingandExecution
A program is loaded in the memory before it can be executed. Operating system provides
the facility to load programs in memory easily and then execute it.
4. Datasecurity
Data is an important partofcomputer system. The operating systemprotectsthe data storedon the
computer from illegal use, modification or deletion.
5. DiskManagement
Operatingsystemmanagesthe diskspace.Itmanagesthestoredfilesandfoldersinaproper way.
6. ProcessManagement
CPUcanperformonetask at onetime. ifthereare many tasks, operating systemdecideswhich task
should get the CPU.
7. DeviceControlling
operating systemalso controlsalldevicesattached to computer.The hardware devices are
controlled with the help of small software called device drivers..
8. Providinginterface
It is used in order that user interface acts with a computer mutually. User interface controls
how youinput dataand instructionandhow information isdisplayedonscreen.Theoperating
system offers two types of the interface to the user:
1. Graphical-lineinterface:It interactswithofvisualenvironmentto communicate
with the computer. It uses windows, icons, menus and other graphicalobjectsto issues
commands.
2. Command-line interface:it provides an interface to communicate with the computer by
typing commands.

3
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

ComputerSystem Architecture
Computersystemcan bedivided intofourcomponentsHardware–provides basic
computing resources
CPU,memory,I/Odevices,Operating system
Controlsandcoordinatesuseofhardwareamongvariousapplicationsandusers
Applicationprograms–definethewaysinwhichthesystemresourcesareusedtosolvethecomputing problems of
the users
Wordprocessors,compilers,webbrowsers,databasesystems,videogames
Users
People,machines,othercomputersFour
Components of a Computer System

Computer architecture means construction/design ofa computer. A computer system may be

organized in different ways. Some computer systems have single processor and others have
multiprocessors. So based on the processors used in computer systems, they are categorized
into the following systems.

1. Single-processorsystem

2. Multiprocessorsystem

3. ClusteredSystems:

1. Single-ProcessorSystems:

Some computers use only one processor such as microcomputers (or personal computers PCs).
On a single-processor system, there is only one CPU that performs all the activities in the
computer system. However, most of these systems have other special purpose processors, such
as I/O processors that move data quickly among different components of the computers. These
processorsexecuteonlya limited systemprograms and do not runtheuser program. Sometimes
4
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

they are managed by the operating system. Similarly, PCs contain a special purpose
microprocessorinthekeyboard,whichconvertsthekeystrokesintocomputercodesto besentto the
CPU.The use ofspecialpurpose microprocessors iscommon inmicrocomputer. But itdoes not
mean that this system is multiprocessor. A system that has only one general-purpose CPU,is
considered as single- processor system.

2. MultiprocessorSystems:

In multiprocessor system, two or more processorsworktogether. Inthis system, multiple programs

(more than one program) are executed on different processors at the same time. This type of
processing isknown as multiprocessing. Some operating systems have featuresof multiprocessing.
UNIX is an example of multiprocessing operating system. Some versions of Microsoft Windows
also support multiprocessing.

Multiprocessor system is also known as parallel system. Mostly the processors of

multiprocessor systemshare the common system bus, clock, memory and peripheral devices.
This system is very fast in data processing.

TypesofMultiprocessorSystems:

Themultiprocessorsystemsarefurtherdividedintotwo types; (i).

Asymmetric multiprocessing system
(ii).Symmetricmultiprocessingsystem

(i) AsymmetricMultiprocessing System(AMS):

The multiprocessing system, in which each processor is assigned a specific task, is known as
Asymmetric Multiprocessing System. For example, one processor is dedicated for handling
user's requests, one processor isdedicated for running application program, and one processor
is dedicated for running image processing and so on. In this system, one processor works as
master processor, while other processors work as slave processors. The master processor
controls the operations of system. It also schedules and distributes tasks among the slave
processors. The slave processors perform the predefined tasks.

(ii) SymmetricMultiprocessing System(SMP):

The multiprocessing system, in which multiple processors worktogether on the same task, is
known as Symmetric Multiprocessing System. In this system, each processor can perform all
typesoftasks.Allprocessorsaretreatedequallyandnomaster-slaverelationshipexists between the
processors.

5
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

For example, different processorsinthesystemcancommunicatewitheachother. Similarly, an I/O

canbe processed onanyprocessor. However, I/O must be controlled to ensure that the data
reaches the appropriate processor. Because all the processors share the same memory, so the
input data given to the processors and their results must be separately controlled. Today all
modern operating systems including Windows and Linux provide support for SMP.

It mustbe noted that in the same computer system, the asymmetric multiprocessing
andsymmetric multiprocessing technique can be used through different operating systems.

ADual-CoreDesign

3. ClusteredSystems:

Clustered system is another formof multiprocessor system. This systemalso contains multiple
processors but it differs from multiprocessor system. The clustered system consists of two or
more individual systems that are coupled together. In clustered system, individual systems (or
clustered computers) share the same storage and are linked together ,via Local Area Network
(LAN).

A layer of cluster software runs on the cluster nodes. Each node can monitor one or more of
the other nodes over the LAN. If the monitored machine fails due to some technical fault (or
due to other reason), the monitoring machine can take ownership of its storage. The
monitoring machine can also restart the applications that were running on the failed machine.
The users of the applications see only an interruption of service.

TypesofClusteredSystems:

Likemultiprocessorsystems,clusteredsystemcanalsobeoftwo types (i).

Asymmetric Clustered System
(ii).SymmetricClustered System
(i). AsymmetricClusteredSystem:
Inasymmetricclusteredsystem,onemachineisinhot-standbymodewhiletheother
6
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

machine is running the application. The hot-standby host machine does nothing. It only
monitors the active server. If the server fails, the hot-standby machine becomes the active
server.
(ii). SymmetricClusteredSystem:
In symmetric clustered system, multiple hosts (machines) run the applications. They also
monitor each other. This mode is more efficient than asymmetric system, because it uses all
the available hardware. This mode is used only if more than one application be available to
run.

Operating System – Structure

OperatingSystemStructure
Multiprogrammingneeded for efficiency
Single user cannot keep CPU and I/O devices busy at all times
Multiprogramming organizesjobs (code and data) so CPU alwayshas one
toExecute A subset of total jobs in system is kept in memory

7
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

8
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

2)Multitasking

9
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Operating-systemOperations

1) Dual-ModeOperation·
In order to ensure the proper execution of the operating system, we must be able to distinguish
between the execution of operating-system code and user defined code. The approach taken by
most computer systems is to provide hardware support that allows us to differentiate among
various modes of execution.

Attheveryleastweneedtwoseparatemodesofoperation.usermodeandkernelmode.
Abit,calledthe modebit isaddedtothehardwareofthecomputerto indicatethecurrent mode: kernel
(0) or user (1).with the mode bit we are able to distinguish between a task that isexecuted
onbehalfofthe operatingsystemandone thatisexecuted onbehalfoftheuser,When

the computer system is executing on behalf of a user application, the system is in user mode.
However, when a user application requests a service from the operating system (via a.. system
call), it must transition from user to kernel mode to fulfill the request.

At system boot time, the hardware starts in kernel mode. The operating system is then loaded
and starts user applications in user mode. Whenever a trap or interrupt occurs, the hardware
switches fromuser mode to kernel mode (that is, changes the state ofthe mode bit to 0). Thus,
whenever the operating system gains control of the computer, it is inkernel mode. The system
always switches to user mode (by setting the mode bit to 1) before passing control to a user
program.
10
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET
The dual mode of operation provides us with the means for protecting the operating system
from errant users-and errant users from one another. We accomplish this protection by
designating some of the machine instructions that may cause harm as privileged instructions.the
hardware allows privileged instructions to beexecuted only inkernel mode. Ifan attempt is made
to execute a privileged instruction in user mode, the hardware does not execute the instruction
but rather treats it as illegal and traps it to the operating system. The instruction to switch to
kernel mode is an example of a privileged instruction. Some other examples include I/0 control
timer management and interrupt management.

11
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

12
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

13
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

14
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Personal-ComputerSystems(PCs)
A personal computer (PC) is a small, relatively inexpensive computer designed for an
individual user. In price, personal computers range anywhere from a few hundred dollars to
thousands of dollars. All are based on the microprocessor technology that enables
manufacturers to put an entire CPU on one chip.
At home, the most popular use for personal computers is for playing games. Businessesuse
personal computers for word processing, accounting, desktop publishing, and for
runningspreadsheet and database managementapplications.

15
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Specialpurposesystems

a) Real-TimeEmbeddedSystems
These devices are found everywhere, fromcar engines and manufacturing robotsto DVDs
and microwave ovens. They tend to have very specific tasks.
Theyhavelittleornouserinterface,preferringtospendtheirtimemonitoringand managing
hardware devices, such as automobile engines and robotic arms.

16
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

b) MultimediaSystems
Mostoperating systemsaredesignedto handleconventionaldatasuchastext files, programs,
word-processing documents, and spreadsheets. However, a recent trend in technology is the
incorporation of multimedia data into computer systems. Multimedia data consist of audio
and video files as well as conventional files. These data differ fromconventional data in that
multimedia data-such as frames of video-must be delivered (streamed) according to certain
time restrictions (for example, 30 frames per second). Multimedia describes a wide range of
applications in popular use today. These include audio files such as MP3, DVD movies,video
conferencing, and short video clips of movie previews or news stories downloadedover the
Internet. Multimedia applications mayalso include live webcasts (broadcasting over the
World Wide Web)

c) HandheldSystems
Handheld Systems include personaldigitalassistants (PDAs, cellular telephones. Developers of
handheld systems and applications face many challenges, most of which are due to the limited
size of such devices. For example, a PDA is typically about 5 inches in height and 3 inches in
width, and it weighs less than one-half pound. Because of their size, most handheld deviceshave
small amounts of memory, slow processors, and small display screens.

17
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

OperatingSystemServices

One set of operating-system services provides functions that are helpful to theuser
Communications–Processes mayexchange information, onthesamecomputerorbetweencomputers
overanetworkCommunicationsmaybe viasharedmemoryorthroughmessagepassing(packetsmoved by the
OS)
Error detection – OS needs to beconstantlyaware of possible errors Mayoccur inthe CPU and
memoryhardware, inI/Odevices, inuserprogramForeachtypeoferror,OSshouldtaketheappropriate action
to ensure correct and consistent computing Debugging facilities can greatly enhance the user’s and
programmer’s abilities to efficiently use the system
AnothersetofOSfunctionsexistsfor ensuringtheefficient operationofthesystemitselfviaresource
Sharing

18
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET
Resourceallocation -Whenmultipleusersormultiple jobsrunningconcurrently, resources must be
allocated to each of them
Manytypesofresources- Some(suchasCPU cycles, mainmemory, and filestorage) mayhavespecial
allocation code, others (such as I/O devices) mayhave general request and release code
Accounting -Tokeeptrackofwhichusersusehow muchandwhatkindsofcomputerresources
Protection and security - The owners of information stored in a multiuser or networked computer
system may want to control use of that information, concurrent processes should not interfere with each
other
Protectioninvolvesensuringthatallaccesstosystemresources iscontrolled
Security ofthe systemfromoutsiders requires user authentication, extends to defending externalI/O
devices from invalid access attempts
Ifa systemisto beprotectedand secure, precautions mustbeinstitutedthroughout it. Achain isonlyas
strong as its weakest link.
UserOperating SystemInterface-CLI
CommandLineInterface(CLI)orcommandinterpreterallowsdirectcommand entry
Sometimesimplementedinkernel,sometimesbysystemsprogram
sometimes multiple flavors implemented – shells
Primarilyfetchesacommandfromuser andexecutesit

UserOperating SystemInterface-GUI

User-friendly desktop metaphor interfaceUsually

mouse, keyboard, and monitor Icons represent
files, programs, actions, etc
Various mouse buttons over objects in the interface cause various actions (provide information,
options, execute function, open directory (known as a folder)
InventedatXeroxPARC
Many systems now include both CLI and GUI
interfaces Microsoft Windows is GUI with CLI
“command” shell
Apple Mac OS X as “Aqua” GUI interface with UNIX kernel underneath and shells
available Solaris is CLI with optional GUI interfaces (Java Desktop, KDE)
SystemCalls

Programming interface to the services provided by the OS

Typically written in a high-level language (C or C++)
Mostly accessed by programs via a high-level Application Program Interface (API) rather than
direct systemcallusenThree most commonAPIsare Win32 API for Windows, POSIXAPI for POSIX-
based systems (including virtually all versions of UNIX, Linux, and Mac OS X), and Java API for the
Java virtual machine (JVM)
WhyuseAPIsratherthansystemcalls?

19
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET
ExampleofSystemCalls

ExampleofStandardAPI
ConsidertheReadFile()functioninthe
Win32API—afunctionforreadingfromafile

Adescriptionoftheparameterspassedto ReadFile()HANDLEfile—the filetoberead

LPVOID buffer—a buffer where the data will be read into and written
fromDWORD bytesToRead—the number ofbytes to be read into the
buffer LPDWORD bytesRead—the number of bytes read during the
last read LPOVERLAPPED ovl—indicates ifoverlapped I/O is being
used
SystemCallImplementation
Typically, anumber associatedwitheachsystemcall
System-callinterfacemaintainsatableindexedaccordingtotheseNumbers
The system call interface invokes intended system call in OS kernel and returns status of the system
call and any return values
The caller need know nothing about how the system call is
implementedJustneedstoobeyAPIandunderstandwhatOSwill
20
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET
doasaresultcallMostdetailsofOSinterfacehiddenfrom programmer by API
Managed by run-time support library (set of functions built into libraries included with compiler)
API – System Call – OS Relationship
StandardCLibraryExample

SystemCallParameterPassing
Often,moreinformationisrequiredthansimplyidentity ofdesiredsystem call
Exact type and amount of information vary according to OS and call Three
general methods used to pass parameters to the
OSSimplest:passtheparametersinregisters
Insomecases,maybemoreparametersthanregisters
Parameters stored ina block, or table, inmemory, and address ofblock passed as a parameter in
a register
ThisapproachtakenbyLinuxandSolaris
Parameters placed, or pushed, onto the stack bythe programand popped off the stack bytheoperating
system
Blockandstackmethodsdonotlimitthe numberorlengthofparametersbeingpassed

21
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET
ParameterPassingviaTable

TypesofSystemCalls
1. Processcontrol
2. Filemanagement
3. Devicemanagement
4. Informationmaintenance
5. Communications
Processcontrol
Arunningneedstohaltitsexecutioneithernormallyor abnormally.
If a system call is made to terminate the running program, a dump of memory is
sometimestaken and an error message generated which can be diagnosed by a debugger
o end,abort
oload,execute
o createprocess,terminate process
ogetprocessattributes,setprocessattributes
owaitfor time
owaitevent,signalevent
oallocateandfreememory
Filemanagement
OSprovidesanAPI tomakethesesystemcallsformanagingfiles
ocreatefile,deletefile
oopen,closefile
oread, write, reposition
ogetandsetfileattributes
Device management
Process requires several resources to execute, if these resources are available, they will be
granted and control retuned to user process. Some are physical such as video card and other
such as file. User program request the device and release when finished
orequestdevice,releasedevice
oread, write, reposition
ogetdeviceattributes, setdeviceattributes
o logicallyattachordetachdevices
22
 OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Informationmaintenance
System calls exist purely for transferring information between the user
programandOS. It canreturninformationaboutthesystem, suchasthenumberofcurrent users, the
versionnumber ofthe operating system, the amount offree memoryor disk space and so on.
o gettimeordate,settimeordate
o getsystemdata,setsystemdata
o getandsetprocess,file,ordevice attributes

Communications
Twocommonmodelsofcommunication
Message-passingmodel,informationisexchangedthroughaninterprocess-
communication facility provided by the OS.
Shared-memory model, processes use map memorysystemcalls to gainaccess to regions of
memory owned by other processes.
ocreate, deletecommunicationconnection
osend,receivemessages
otransferstatus information
oattachanddetachremotedevices

ExamplesofWindowsandUnixSystemCalls

23
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

MS-DOS execution

(a) At system startup (b)runninga

program FreeBSD Running Multiple Programs

24
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

SystemPrograms
Systemprogramsprovideaconvenient environment forprogramdevelopment andexecution. Thecan be
divided into:
Filemanipulation
Statusinformation
File modification
Programminglanguagesupport
Program loading and execution
Communications
Applicationprograms

Most users’ view of the operation system is defined by system programs, not the actual
systemcalls provide a convenient environment for programdevelopment and execution
Someofthemare simplyuserinterfacesto systemcalls;othersare considerablymore complex
Filemanagement-Create,delete,copy,rename,print,dump,list,andgenerallymanipulatefilesand directories
 Statusinformation
Some ask the system for info - date, time, amount of available memory, disk space, number of users
Others provide detailed performance, logging, and debugging information
Typically, these programs format and print the output to the terminal or other output
devicesSome systems implement a registry - used to store and retrieve configuration
information
 Filemodification
Texteditorstocreateandmodifyfiles
Special commands to search contents of files or perform transformations of the text
Programming-language support - Compilers, assemblers, debuggers and interpreters sometimes
provided
Program loading and execution- Absolute loaders, relocatable loaders, linkage editors, andoverlay-
loaders, debugging systems for higher-level and machinelanguage

25
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET
Communications- Providethe mechanismforcreating virtualconnectionsamongprocesses, users,and
computer systems
Allowuserstosendmessagestooneanother’sscreens,browsewebpages,sendelectronic-mail messages, log
in remotely, transfer files from one machine toanother

OperatingSystemDesign andImplementation
Design and Implementation ofOS not “solvable”, but some approaches have proven successful
Internal structure of different Operating Systems can vary widely
StartbydefininggoalsandspecificationsAffectedby
choice of hardware, type of system User goals and
System goals
Usergoals–operatingsystemshouldbeconvenienttouse,easytolearn,reliable, safe, andfast
System goals–operatingsystem shouldbeeasytodesign,implement,andmaintain,aswell asflexible, reliable,
error-free, and efficient
Importantprincipletoseparate
Policy:Whatwillbe done?
Mechanism:Howtodoit?
Mechanismsdeterminehowtodosomething,policiesdecidewhatwillbe done
Theseparationof policyfrommechanismisaveryimportantprinciple,itallowsmaximumflexibilityif policy
decisions are to be changed later
SimpleStructure
MS-DOS – written to provide the most functionality in the least space Not divided
intomodules
AlthoughMS-DOShassomestructure,itsinterfacesandlevelsofFunctionalityarenotwellseparated

26
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

MS-DOSLayerStructure

The operating system isdivided into a numberoflayers(levels), each built ontopoflower layers.The
bottom layer(layer 0), isthe hardware; the highest (layer N) isthe userinterface.
With modularity, layers are selected such that each uses functions (operations) and services of
only lower-level layers
TraditionalUNIXSystemStructure

UNIX

UNIX–limited byhardware functionality,theoriginalUNIXoperatingsystemhad limitedstructuring. The

UNIX OS consists of two separable parts
Systemsprograms Th
e kernel
Consistsofeverythingbelowthesystem-callinterfaceandabovethephysicalhardware Provides the
file system, CPU scheduling, memory management, and other operating-system

27
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET
functions;alargenumberoffunctionsforonelevel Layered
Operating System

MicrokernelSystemStructure
Movesasmuch fromthekernelinto“user”space
Communicationtakesplacebetweenusermodulesusingmessagepassing Benefits:
Easiertoextenda microkernel
Easier to port the operating system to new architectures More reliable (less codeis
running in kernel mode)
Moresecure
Detriments:
Performanceoverheadofuserspacetokernelspacecommunication
MacOS X Structure

28
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Modules

Most modern operating systems implement kernel modules

Uses object-oriented approach
Eachcorecomponentisseparate
Each talks to the others over known interfacesEach
is loadable as needed within the kernel Overall,
similar to layers but with more flexible

SolarisModularApproach

VirtualMachines
Avirtualmachinetakesthelayeredapproachtoitslogicalconclusion.Ittreats hardwareand the operating system
kernel as though they were allhardware
Avirtualmachineprovidesaninterfaceidenticaltotheunderlyingbarehardware
Theoperatingsystem hostcreatestheillusion thataprocesshasitsown processorand(virtual memory) Each guest
provided with a (virtual) copy of underlying computer
VirtualMachines HistoryandBenefits
First appearedcommerciallyinIBMmainframesin1972
Fundamentally, multipleexecutionenvironments(differentoperatingsystems)cansharethesamehardware
Protect from each other
Somesharing offilecanbepermitted,controlled
Commutate with each other, other physical systems via networking
Useful for development, testing
Consolidationofmanylow-resourceusesystemsontofewerbusiersystems
“OpenVirtualMachineFormat”,standardformatofvirtualmachines,allowsaVMtorunwithinmany different
virtual machine (host) platforms

29
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Para-virtualization
Presentsguestwithsystemsimilarbutnotidenticaltohardware Guest
must be modified to run on par virtualized hardware
GuestcanbeanOS,orinthecaseofSolaris10applicationsrunningincontainers Solaris 10 with
Two Containers

30
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET
VMwareArchitecture

TheJava VirtualMachine

Operating-SystemDebugging

Debugging is finding and fixing errors, or bugs

generate log files containing error information
Failure of an application can generate core dump file capturing memory of the process
Operating system failure can generate crash dump file containing kernel memory Beyond
crashes, performance tuning can optimize system performance
Kernighan’s Law: “Debugging is twice as hard as writing the code in the rst place. Therefore, if you
write the code as cleverlyas possible, you are, bydefinition, not smart enoughto debug it.”
DTrace tool in Solaris, FreeBSD, Mac OS X allows live instrumentation on production systems
Probesfirewhencodeisexecuted,capturingstatedataandsending it toconsumersofthoseprobes

31
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET
Process
Aprocessis aprogramatthetimeofexecution.
DifferencesbetweenProcessand Program

Process Program
Processisadynamicobject Programisastaticobject

Process is sequence of instruction Programisasequence ofinstructions

execution
Processloaded into mainmemory Programloadedintosecondarystorage
devices
Timespanofprocessislimited Timespanofprogramisunlimited
Processisaactiveentity Programisapassive entity

Process States

When a process executed, it changes the state, generally the state of process is determined by
the current activityof the process. Each process may be in one of the following states:
1. New :Theprocessisbeingcreated.
2. Running :Theprocessisbeingexecuted.
3. Waiting :Theprocess iswaitingforsomeeventtooccur.
4. Ready :Theprocessiswaitingtobeassignedtoaprocessor.
5. Terminated:TheProcesshasfinishedexecution.
Onlyoneprocesscanberunninginanyprocessoratany time,Butmanyprocessmaybein ready and
waiting states. The ready processes are loaded into a “ready queue”.
Diagramofprocessstate

32
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

a) New ->Ready : OScreatesprocessandpreparesthe

process to be executed,thenOSmoved the process into readyqueue.
b) Ready->Running :OSselectsoneoftheJobs fromreadyQueueandmovethemfrom ready
to Running.

c) Running->Terminated: When the Execution of a process has Completed,

OSterminatesthatprocess from running state. Sometimes OS terminates the process for
someother reasons including Time exceeded, memory unavailable, access violation,
protection Error, I/O failure and soon.
d) Running->Ready:Whenthetimeslot oftheprocessor expired(or)Ifthe
processorreceivedanyinterrupt signal, the OS shifted Running -> ReadyState.

e) Running->Waiting:Aprocessisputintothewaitingstate,iftheprocessneedan event occur

(or) an I/O Devicerequire.
f) Waiting->Ready : A processin the waiting state ismoved to ready
state whenthe eventforwhichit has beenCompleted.
ProcessControlBlock:

Eachprocessisrepresented intheoperating SystembyaProcessControlBlock.

ItisalsocalledTaskControlBlock.Itcontainsmanypiecesofinformationassociatedwithaspecific Process.

ProcessState

ProgramCounter

CPURegisters

CPUScheduling Information

Memory–ManagementInformation

AccountingInformation

I/OStatusInformation

ProcessControlBlock
1. ProcessState :TheStatemaybenew,ready,running,andwaiting,Terminated…
2. ProgramCounter :indicatestheAddressofthenextInstructionto beexecuted.
3. CPUregisters :registersincludeaccumulators,stackpointers,
General purpose Registers….

33
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

4. CPU-SchedulingInfo : includes a process pointer, pointers to

schedulingQueues,other scheduling parametersetc.
5. Memory managementInfo: includespagetables, segmentationtables, value of
base and limit registers.
6. AccountingInformation: includesamountofCPUused,timelimits,Jobs(or)Processnumbers.
7. I/O StatusInformation: Includes the list ofI/O Devices Allocated to theprocesses, list ofopen
files.

Threads:

Aprocess isdivide into numberoflight weight process, each light weight processissaid to be a
Thread. The Thread has a program counter (Keeps track of which instruction to execute next),
registers (holds its current working variables), stack (execution History).
Thread States:

1. bornState : Athread isjustcreated.

2. readystate :ThethreadiswaitingforCPU.
3. running :Systemassignsthe processortothethread.
4. sleep :Asleeping threadbecomesreadyafterthedesignatedsleeptimeexpires.
5. dead :TheExecutionofthethreadfinished.

Eg:Wordprocessor.
Typing,Formatting,Spellcheck,savingarethreads.
DifferencesbetweenProcessand Thread

Process Thread
Processtakesmoretimeto create. Thread takeslesstimeto create.
ittakesmoretimetocompleteexecution& Lesstimeto terminate.
terminate.
Execution isveryslow. Executionisveryfast.
It takes more time to switch b/w two Ittakeslesstimetoswitchb/wtwo
processes. threads.
Communicationb/wtwoprocessesisdifficult . Communicationb/wtwothreadsis
easy.
Processcan’tsharethesame memoryarea. Threadscansharesamememoryarea.
Systemcallsarerequestedtocommunicate Systemcallsare notrequired.
each other.
Processislooselycoupled. Threadsaretightlycoupled.
Itrequiresmoreresourcesto execute. Requiresfewresourcestoexecute.

34
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Multithreading

AprocessisdividedintonumberofsmallertaskseachtaskiscalledaThread.Numberof Threads with

in a Process execute at a time is called Multithreading.
If a program, is multithreaded, even when some portion of it is blocked, the whole program is
not blocked.The rest of the program continues working If multiple CPU’s are available.
Multithreading gives best performance.If we have only a single thread, number of CPU’s
available, No performance benefits achieved.
 Processcreationisheavy-weightwhilethreadcreationis light-weight
Can simplify code, increase efficiency

Kernelsaregenerallymultithreaded
CODE-Contains instruction
DATA-holdsglobalvariableFILES-
openingandclosingfiles
REGISTER-containinformationaboutCPUstate
STACK-parameters, local variables, functions
Types Of Threads:

1) User Threads : Thread creation, scheduling, management happen in user space by

ThreadLibrary. user threads are fasterto create and manage. Ifa user thread performs a system
call, which blocks it, allthe otherthreads inthat process one also automaticallyblocked, whole
process is blocked.

Advantages
 ThreadswitchingdoesnotrequireKernelmodeprivileges.
 Userlevelthreadcanrunonanyoperatingsystem.
 Schedulingcanbeapplicationspecificintheuserlevelthread.
 Userlevelthreadsarefasttocreateand manage.
35
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Disadvantages

 Inatypicaloperatingsystem,mostsystemcallsareblocking.
2) Kernel Threads: kernel creates, schedules, manages these threads .these threads are
 Multithreadedapplicationcannottakeadvantageofmultiprocessing.
slower, manage. Ifone thread in a process blocked, over allprocess need not be blocked.

Advantages
Disadvantages
 Kernelcansimultaneouslyschedulemultiplethreadsfromthesameprocess onmultiple
processes.
 Ifonethreadinaprocessis blocked, theKernelcanscheduleanotherthreadofthesameprocess.
Kernelthreadsaregenerallyslowertocreateand managethantheuserthreads.
 Kernelroutinesthemselvescanmultithreaded.
Transfer ofcontrolfromonethreadtoanother withinsameprocess requiresa modeswitchto the Kernel.

 MultithreadingModels
Manytomanyrelationship.
 Manytoonerelationship.
 Some
Onetoonerelationship.
operating systemprovides a combined user level thread and Kernel level thread facility. Solaris
isa good example of this combined approach. In a combined system, multiple threads within the same
ManytoManyModel
applicationcanruninparallel onmultipleprocessorsanda blockingsystemcallneed not blocktheentire
process. Multithreading models are three types
Inthis model,manyuser levelthreads multiplexestotheKernelthread ofsmaller orequalnumbers.The number of
Kernel threads maybespecific to either a particular application ora particularmachine.
Followingdiagramshowsthemanytomanymodel.In thismodel,developerscancreateasmanyuser 36
threadsasnecessaryandthecorrespondingKernelthreadscanruninparallelsonamultiprocessor.
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

ManytoOneModel

Manytoonemodelmaps manyuser levelthreadstooneKernellevelthread.Threadmanagementis done in user

space. When thread makes a blocking system call, the entire process will be blocks. Only one
threadcanaccess theKernelatatime,so multiplethreads areunabletoruninparallel on multiprocessors.

Iftheuser levelthreadlibraries areimplementedintheoperatingsysteminsucha waythat systemdoes not

support them then Kernel threads use the many to one relationship modes.

OnetoOneModel

Thereis onetoonerelationship ofuser levelthreadtothekernellevelthread.This modelprovides more

concurrency than the many to one model. It also another thread to run when a thread makes a blocking
system call. It support multiple thread to execute in parallel on microprocessors.

Disadvantageofthis modelisthatcreatinguser threadrequiresthecorrespondingKernelthread.OS/2,

windows NT and windows 2000 use one to one relationship model.

37
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

38
 OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

UNIT-II
Process Scheduling: Foundation and Scheduling objectives, Types of Schedulers, Scheduling criteria:
CPU utilization, Throughput, Turnaround Time, Waiting Time, Response Time; Scheduling algorithms:
Pre-emptiveandNonpre-emptive, FCFS, SJF, RR;Multiprocessor scheduling:RealTimescheduling:RM and
EDF.
Inter-processCommunication: CriticalSection, Race Conditions, MutualExclusion, Hardware Solution,
Strict Alternation, Peterson’s Solution, The Producer/Consumer Problem, Semaphores, Event Counters,
Monitors, Message Passing, Classical IPC Problems: Reader’s & Writer Problem, Dinning Philosopher
Problem etc.

PROCESS SCHEDULING:

CPUisalwaysbusyinMultiprogramming. BecauseCPUswitches fromonejobtoanotherjob. Butin

simplecomputersCPUsitidleuntiltheI/Orequestgranted.
scheduling is a important OS function. All resources are scheduled before use.(cpu,
memory, devices…..)
Process scheduling is an essential part of a Multiprogramming operating systems. Such
operatingsystemsallow morethanoneprocesstobe loaded into theexecutable memoryat a
time and the loaded process shares the CPU using time multiplexing
. SchedulingObjectives
Maximizethroughput.
Maximizenumberofusersreceivingacceptableresponsetimes. Be
predictable.
Balanceresourceuse.
Avoidindefinitepostponement.
Enforce Priorities.
Givepreferencetoprocessesholdingkeyresources

SCHEDULING QUEUES: people live in rooms. Process are present in rooms knows
as queues. There are 3types
1. job queue: when processes enter the system, they are put into a job queue, which
consistsallprocesses inthe system. Processes inthe jobqueue reside onmassstorageand await the
allocation of main memory.
2. ready queue:if a processis presentin mainmemory andis ready to be allocated to cpu
forexecution, is kept in readyqueue.
3. devicequeue:ifaprocessispresentinwaitingstate(or)waitingforani/oeventto complete is
said to bein device queue.(or)
Theprocesseswaiting foraparticularI/Odeviceiscalleddevicequeue.

39
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Schedulers:Thereare3 schedulers

1. Longtermscheduler.
2. Mediumterm scheduler
3. Shorttermscheduler.

Schedulerduties:

 Maintainsthequeue.
 SelecttheprocessfromqueuesassigntoCPU.
Typesof schedulers

1. Long term scheduler:

select the jobs from the job pool and loaded these jobs into main memory (ready
queue).Long term scheduler is also called job scheduler.
2. Shorttermscheduler:
select theprocessfromreadyqueue,and allocatesitto the cpu.
If a process requires an I/Odevice,whichis not presentavailable then process enters device
queue.
shorttermschedulermaintainsreadyqueue,devicequeue.Alsocalledascpuscheduler.
3. Medium term scheduler: if process request an I/O device in the middle of the
execution, then the process removed fromthe main memory and loaded into the waiting queue.
When the I/O operation completed, then the job moved from waiting queue to ready queue.
These two operations performed by medium term scheduler.

40
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Context Switch: Assume, main memory contains more than one process. If cpu is executing a process, if
time expires or if a high priority process enters into main memory, then the scheduler saves information
about current process in the PCB and switches to execute the another process. The concept of moving CPU
by scheduler from one process to other process is known as context switch.
Non-Preemptive Scheduling: CPU is assigned toone process, CPU do not release untilthe competitionof
that process. The CPU will assigned to some other process only after the previous process has finished.
Preemptive scheduling: here CPU can release the processes even in the middle of the
execution. CPU received a signal from process p2. OS compares the priorities of p1 ,p2. If
p1>p2, CPU continues the execution of p1. If p1<p2 CPU preempt p1 and assigned to p2.
Dispatcher: The main job of dispatcher is switching the cpu from one process to another
process. Dispatcher connects the cpu to the process selected by the short term scheduler.
Dispatcher latency: The time it takes by the dispatcher to stop one process and start another
processisknownasdispatcher latency.Ifthedispatcher latencyisincreasing,thenthedegreeof
multiprogramming decreases.
SCHEDULINGCRITERIA:

1. Throughput:howmanyjobsarecompletedbythecpuwithinatimeperiod.
2. Turn around time : The time interval between the submission ofthe process
and time of the completion is turn around time.
TAT = Waiting time in ready queue + executing time + waiting time in waiting queue for
I/O.
3. Waitingtime:Thetimespentbytheprocesstowaitfor cputo beallocated.
4. Responsetime:Timedurationbetweenthesubmissionandfirstresponse.
5. CpuUtilization:CPUiscostlydevice, it mustbekeptasbusyaspossible. Eg:
CPU efficiency is 90% means it is busy for 90 units, 10 units idle.
CPUSCHEDULINGALGORITHMS:

1. First come First served scheduling: (FCFS): The process that request the CPU
first is holds the cpu first. If a process request the cpu then it is loaded into the ready queue,
connect CPU to that process.
Consider the following set of processes that arrive at time 0, the length of the cpu burst time
given in milli seconds.
bursttimeisthetime,requiredthecputoexecute thatjob,itisinmilliseconds.

Process Bursttime(milliseconds)
P1 5
P2 24
P3 16
P4 10
P5 3

41
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Averageturnaroundtime:

Turnaroundtime=waitingtime+bursttime

Turnaroundtimeforp1=0+5=5.
Turn around time for
p2=5+24=29 Turn around time
for p3=29+16=45 Turn around
time for p4=45+10=55 Turn
around time for p5= 55+3=58
Averageturnaroundtime= (5+29++45+55+58/5)=187/5=37.5millisecounds

Averagewaitingtime:

waitingtime=startingtime-arrivaltime
Waitingtimeforp1=0
Waiting time for p2=5-0=5
Waiting time for p3=29-0=29
Waiting time for p4=45-0=45
Waiting time for p5=55-0=55
Averagewaitingtime=0+5+29+45+55/5 =125/5=25 ms.

AverageResponseTime:

Formula :FirstResponse -Arrival

Time Response Time for P1 =0
Response Time for P2 => 5-0 = 5
Response Time for P3 => 29-0 = 29
Response Time for P4 => 45-0 = 45
Response Time for P5 => 55-0 = 55
AverageResponseTime=>(0+5+29+45+55)/5 =>25ms

42
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

1) FirstComeFirstServe:

ItisNonPrimitiveSchedulingAlgorithm.

PROCESS BURST ARRIVAL

TIME TIME
P1 3 0

P2 6 2

P3 4 4

P4 5 6

P5 2 8

Processarrived intheorderP1,P2,P3,P4,P5. P1
arrived at 0 ms.
P2 arrived at 2 ms.
P3 arrived at 4 ms.
P4 arrived at 6 ms.
P5 arrived at 8 ms.

AverageTurnAroundTime
Formula:TurnaroundTime=:waitingtime+bursttime Turn
Around Time for P1 => 0+3= 3
Turn Around Time for P2 => 1+6 = 7
Turn Around Time for P3 => 5+4 = 9
TurnAround Time for P4 => 7+ 5=12
TurnAroundTimeforP5=>2+10=12
AverageTurnAround Time=>(3+7+9+12+12)/5=>43/5 =8.50 ms.
AverageResponseTime:
Formula:ResponseTime=FirstResponse-ArrivalTime
Response Time of P1 = 0
Response Time of P2 => 3-2 = 1
Response Time of P3 => 9-4 = 5
Response Time ofP4 => 13-6 = 7
ResponseTimeofP5=>18-8=10
AverageResponseTime=>(0+1+5+7+10 )/5=>23/5 =4.6ms
Advantages:EasytoImplement,Simple.
43
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Disadvantage:Averagewaitingtimeisveryhigh.
2) ShortestJobFirst Scheduling(SJF):

Which processhavingthesmallestCPUbursttime,CPUisassignedto thatprocess.If two process

having the same CPU burst time, FCFS is used.

PROCESS CPUBURSTTIME

P1 5

P2 24

P3 16

P4 10

P5 3

P5 having the leastCPU burst time ( 3ms ). CPU assigned to that( P5 ). After completion ofP5
short term scheduler search for nest ( P1 ).......
AverageWaiting Time:

Formula=StaringTime-ArrivalTime waiting
Time for P1 => 3-0 = 3
waitingTimeforP2=>34-0=34
waiting Time for P3 =>18-0 =18
waiting Time for P4 =>8-0=8
waiting time for P5=0
Averagewaiting time=>(3+34+18+8+0)/5 =>63/5 =12.6 ms

AverageTurnAroundTime:

Formula=waitingTime+burstTime

TurnAroundTimeforP1=>3+5=8
Turn Around for P2 => 34+24 =58
Turn Around for P3 => 18+16 = 34
44
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

TurnAroundTimeforP4=>8+10=18 Turn
Around Time for P5 => 0+3 = 3
AverageTurnaroundtime => (8+58+34+18+3)/5=> 121/5=24.2ms
AverageResponseTime:

Formula:FirstResponse-ArrivalTime

First Response time for P1 =>3-0 = 3

FirstResponsetimeforP2=>34-0=34
FirstResponsetimeforP3=>18-0=18 First
Response time for P4 => 8-0 = 8
FirstResponsetimeforP5=0
Average Response Time => ( 3+34+18+8+0 )/5 => 63/5 = 12.6 ms
SJF is Non primitive scheduling algorithm
Advantages : Least average waiting time
LeastaverageturnaroundtimeLeast
average response time
Averagewaitingtime( FCFS)=25 ms
Averagewaitingtime(SJF)=12.6ms50%timesavedinSJF.
Disadvantages:
 KnowingthelengthofthenextCPUbursttimeisdifficult.
 Aging(BigJobsarewaitingforlongtimeforCPU)

3) ShortestRemainingTimeFirst( SRTF);

Thisisprimitivescheduling algorithm.

Short termscheduler always choosesthe process that has termshortest remaining time. Whena
new process joins the ready queue , short term scheduler compare the remaining time of
executing process and new process. If the new process has the least CPU burst time, The
scheduler selects that job and connect to CPU. Otherwise continue the old process.

PROCESS BURSTTIME ARRIVALTIME

P1 3 0
P2 6 2
P3 4 4
P4 5 6
P5 2 8

45
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

P1 arrives at time 0, P1 executing First ,P2arrives attime 2. Compare P1 remaining time and P2 ( 3-2 =
1)and6. So,continueP1after P1,executingP2,attime4,P3arrives,compareP2remainingtime(6-1=5
) and 4 ( 4<5 ) .So, executing P3 at time 6, P4 arrives. Compare P3 remaining time and P4 ( 4-
2=2 ) and 5 (2<5 ). So, continue P3 , after P3, ready queue consisting P5 is the least out ofthree.
So execute P5, next P2, P4.
FORMULA :Finish time - Arrival
Time Finish Time for P1 => 3-0 = 3
Finish Time for P2 => 15-2 = 13
Finish Time for P3 => 8-4 =4
Finish Time for P4 => 20-6 = 14
Finish Time for P5 => 10-8 = 2

AverageTurnaroundtime =>36/5= 7.2ms.

4)ROUNDROBINSCHEDULINGALGORITHM:

It is designed especially for time sharing systems. Here CPU switches between the processes.
Whenthe time quantumexpired, the CPU switched to another job. Asmallunit oftime, called a
time quantum or time slice. A time quantum is generally from 10 to 100 ms. The time quantum
is generally depending on OS. Here ready queue is a circular queue. CPU scheduler picks the
first process from ready queue, sets timer to interrupt after one time quantum and dispatches
the process.

PROCESS BURSTTIME

P1 30

P2 6

P3 8

46
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

AVERAGEWAITINGTIME:

Waitingtimefor P1 =>0+(15-5)+(24-20) =>0+10+4=14

Waitingtimefor P2 =>5+(20-10)=>5+10=15
WaitingtimeforP3=>10+(21-15)=>10+6=16 Average waiting
time => (14+15+16)/3 = 15 ms.

AVERAGETURNAROUNDTIME:
FORMULA:Turnaroundtime=waitingtime+burst Time Turn
around time for P1 => 14+30 =44
Turnaroundtime forP2=>15+6=21
TurnaroundtimeforP3=> 16+8= 24
Averageturnaroundtime => (44+21+24)/3= 29.66ms

5) PRIORITYSCHEDULING:

PROCESS BURST PRIORITY

TIME
P1 6 2

P2 12 4

P3 1 5

P4 3 1

P5 4 3

P4hasthehighestpriority.AllocatetheCPUtoprocessP4firstnextP1,P5,P2,P3.

AVERAGEWAITINGTIME:

Waiting time for P1 => 3-0 =3

Waiting time for P2 => 13-0 = 13
Waiting time for P3 => 25-0 = 25
Waiting time for P4 => 0
WaitingtimeforP5=>9-0=9

Averagewaiting time=>(3+13+25+0+9 )/5=10ms

47
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

AVERAGETURNAROUNDTIME:

Turn around time for P1 =>3+6 = 9

Turnaroundtime forP2=>13+12=25
Turnaroundtime for P3=> 25+1 = 26
TurnaroundtimeforP4=>0+3=3 Turn
around time for P5 => 9+4 = 13

AverageTurnaroundtime=> (9+25+26+3+13 )/5= 15.2ms

Disadvantage:Starvation

Starvationmeansonlyhighpriorityprocessareexecuting,butlowpriority process are

waiting for the CPU for the longest period of the time.

Multiple–processorscheduling:
When multiple processes are available, thenthe scheduling getsmore complicated,
because there is more than one CPU which must be kept busy and in effective use
at all times.
Load sharing resolves around balancing the load between multiple processors.
Multi processor systems may be heterogeneous (It contains different kinds of
CPU’s) ( or ) Homogeneous(all the same kind of CPU).
1) Approaches to multiple-processorscheduling
a)Asymmetric multiprocessing
One processoris themaster,controlling all activities and running all kernel
code,while the other runs only user code.
b)Symmetricmultiprocessing:
Each processor schedules its own job. Each processor may have its own private queue of ready
processes.

2) ProcessorAffinity
Successive memory accesses by the process are often satisfied in cache memory.
what happens if the process migrates to another processor. the contents of cache
memory must be invalidated for the first processor, cache for the second processor
must be repopulated. Most Symmetric multi processor systems try to avoid
migration of processes from one processor to another processor, keep a process
running on the same processor. This is called processor affinity.
a) Soft affinity:
Soft affinity occurs when the system attempts to keep processes on the same
processor but makes no guarantees.

48
 OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

b) Hardaffinity:
Processspecifiesthatitisnottobemovedbetweenprocessors.
3) Loadbalancing:
Oneprocessorwontbesittingidlewhileanotherisoverloaded.Balancing can be
achived through push migration or pull migration.

Pushmigration:
Push migration involves a separate process that runs periodically(e.g every 200 ms)
and moves processes from heavily loaded processors onto less loaded processors.
Pullmigration:
Pullmigrationinvolvesidleprocessorstakingprocessesfromthereadyqueuesoftheother processors.

Realtime scheduling:
Realtime scheduling is generallyused in the case of multimedia operating systems.
Here multiple processescompete for the CPU. How to schedule processes A,B,C so
that each one meets its deadlines. The general tendency is to make them pre-
emptable, so that a process in danger of missing its deadline can preempt another
process. When this process sends its frame, the preempted process can continuefrom
where it had left off. Here throughput is not so significant. Important is that tasks
start and end as per their deadlines.
RATEMONOTONIC(RM)SCHEDULINGALGORITHM
Rate monotonic scheduling Algorithmworksonthe principle ofpreemption. Preemptionoccurs on
a given processor when higher prioritytask blocked lower prioritytask fromexecution. This
blocking occurs due to priority level of different tasks in a given task set.rate monotonic is a
preemptive algorithm which means if a task with shorter period comes during execution it will
gain a higher priorityand can block or preemptive currentlyrunning tasks. In RM priorities are
assigned according totime period. Priorityofa task is inverselyproportionalto itstimer period.
Task with lowest time period has highest priority and the task with highest period will have
lowest priority.
Forexample,wehaveataskset that consistsofthreetasksas follows

Tasks Executiontime(Ci) Timeperiod(Ti)

T1 0.5 3

T2 1 4

T3 2 6

49
 OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Table1.Task set
U= 0.5/3+1/4+2/6=0.167+ 0.25+0.333=0.75
As processor utilization is less than 1or 100% so task set is schedulable and it also satisfies the
aboveequation of rate monotonic scheduling algorithm.

Figure1.RMschedulingofTasksetintable1.
Atasksetgivenintable1itRMschedulingisgiveninfigure1. Theexplanationofaboveisasfollows
1. According to RM scheduling algorithm task with shorter period has higher priority so T1 has
high priority, T2 has intermediate priority and T3 has lowest priority. At t=0 all the tasks are
released. Now T1 has highest priority so it executes first till t=0.5.
2. At t=0.5taskT2has higher prioritythanT3 so it executes first forone-time units tillt=1.5. After its
completion only one task is remained in the system that is T3, so it starts its execution and
executes till t=3.
3. At t=3 T1 releases, as it has higher priority than T3 so it preempts or blocks T3 and starts it
execution till t=3.5. After that the remaining part of T3 executes.
4. At t=4 T2 releases and completes it execution as there is no task running in the system at this
time.
5. At t=6 both T1 and T3 are released at the same time but T1 has higher priority due to shorter
period so it preempts T3 and executes tillt=6.5, after that T3 startsrunning and executes tillt=8.
6. Att=8 T2withhigherprioritythanT3releasessoitpreemptsT3and startsits execution.
7. At t=9 T1 is released again and it preempts T3 and executes first and at t=9.5 T3 executes its
remaining part. Similarly, the execution goes on.

EarliestDeadlineFirst(EDF)SchedulerAlgorithm
The EDF is a dynamic algorithm, Job priorities are re-evaluated at everydecision point, this re-
evaluationis basedonrelativedeadlineofa jobortask,theclosertothedeadline, thehigherthepriority. The EDF
has the following advantages:
1. Veryflexible(arrivaltimesanddeadlinesdonotneedtobeknownbeforeimplementation).
2. Moderatecomplexity.
3. Abletohandleaperiodicjobs.
TheEDFhasthefollowing disadvantages:
1. Optimallyrequirespre-emptivejobs.
2. Notoptimalonseveralprocessors.
3. Difficulttoverify.

50
 OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Example
Considerthefollowingtaskset inTable1. PrepresentsthePeriod, etheExecutiontimeandDstands for the
Deadline. Assume that the job priorities are re-evaluated at the release and deadline ofa job.

P e D

T1 2 0.5 2

T2 4 1 4

T3 5 1.5 5

Solution

Markalldeadlinesrelated to allthetasks
 First markalldeadlinesrelatedtothetasksasshowninFig. 1.T1,T2andT3arerepresented with
Red, Green and Blue colour respectively. The schedule is from0 – 20ms as shown.
 At T =0,T1hastheclosestdeadline, soscheduleT1.
 At T=0.5,T1iscompleted,itsnext releasetime isat 2ms.T2iscloserto itsdeadlineso T2is
scheduled next and executes for 1s.
 At T=1.5,T2jobiscompleted.T3isnext because it isclosertoitsdeadlinewhileT2hasnot been
released.
 At T=2,anew instanceofT1isreleased,therefore,T3isinterruptedandhas1msleft to complete
execution. T1 executes
 AtT=2.5,Theonlyready jobisT3which isscheduleduntilcompletion.
 AtT=4,anewinstanceofT1isreleasedwhichexecutesfor0.5ms.
 AtT =4.5,T1is nowcompleted, soT2isnowthetaskclosesttoitsdeadlineandisscheduled.
 AtT=5.5,T3isscheduledbutispre-emptedatT=6sorunsfor 0.5ms
 AtT=6,anewinstanceofT1 isreleasedand thereforescheduled.
 At T=6.5,T3isclosestto itsdeadlinebecauseT1andT3havenotbeenreleased.SoT3is allowed to
complete its execution which is 1ms.
 AtT=8,anewinstanceofT1isreleasedandisscheduled.
 At T=8.5,T2isthetaskhavingtheclosest deadlineandso isscheduledto runforitsexecution time.
 AtT=10,thenextreleaseofT1is scheduled.

51
 OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

 At T=10.5,thenext jobwiththeclosest deadline isT3becausethenext T2jobwillbereleased at T =

12. So T3 is scheduled until completion.
 AtT=12,thenextreleaseofT1is scheduled.
 AtT =12.5,T2isscheduledas itisthejobwiththeclosestdeadline.
 AtT=14,thenextreleaseofT1is scheduled.
 At T=15,thenext releaseofT3 isscheduledbecause it isnowthejobwiththeclosest deadline because
the next release of T1 and T2 is at 16ms. T3 runs for 1ms.
 At T=16,T3ispre=emptedbecauseanewreleaseofT1whichhas theclosest deadline isnow
available.
 T=16.5,T2isthe jobwiththeclosest deadline,soit isscheduled forthedurationofits
execution time.
 At T=17.5,sinceT1andT2havecompleted,T3resumesexecutionto completeitstaskwhich ran for
only1ms the last time. T3 completes execution at T = 18.
 AtT=18,anewinstanceofT1isreleasedandscheduledtorunfor itsentireexecutiontime.
 AtT =18.5,nojobisreleasedyetbecauseanewreleaseofT1, T2andT3areat20ms.
 Fig. 2showstheEDF schedulefromT= 0toT= 20.
 .

InterProcess communication:

Processsynchronization refersto the idea that multiple processesare to joinup or

handshake at a certain point, in order to reach an agreement or commit to a certain
sequence of action. Coordination of simultaneous processes to complete a task is
known as process synchronization.
Thecriticalsectionproblem
Consider a system, assume that it consisting ofn processes. Eachprocess having a
segment of code. This segment of code is said to be critical section.
E.G:RailwayReservationSystem.
Two persons from different stations want to reserve their tickets, the train number,
destination is common, the two persons try to get the reservation at the same time.
Unfortunately, the available berths are only one; both are trying for that berth.
It is also called the critical section problem. Solution is when one process is
executing in its critical section, no other process is to be allowed to execute in
its critical section.

52
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

The critical section problem is to design a protocol that the processes can use to
cooperate. Each process must request permission to enter its critical section. The
section of code implementing this request is the entry section. The critical section
may be followed byan exit section. The remaining codeis the remaindersection.

A solution to the critical section problem must satisfy the following 3

requirements: 1.mutual exclusion:
Onlyoneprocess canexecute theircriticalsectionatanytime.
2. Progress:
Whennoprocessisexecutingacritical sectionforadata,oneof theprocesses wishing to
enter a critical section for data will be granted entry.
3. Boundedwait:
No processshouldwaitforaresourceforinfiniteamountoftime.

Criticalsection:
Theportioninanyprogramthataccessesasharedresourceiscalledascriticalsection(or) critical region.

Peterson’ssolution:
Peterson solution is one of the solutions to critical section problem involving two
processes. This solutionstatesthat whenone process isexecuting itscriticalsection
then the other process executes the rest of the code and vice versa.
Petersonsolutionrequirestwoshared data items:
1) turn: indicateswhoseturnit isto enter
into the critical section. If turn == i ,then
process i is allowed into theircritical section.
2) flag:indicateswhenaprocesswantstoenterintocriticalsection.when

53
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

processiwantstoentertheircriticalsection,itsetsflag[i]totrue.
do {flag[i] = TRUE; turn = j;
while(flag[j]&&turn==j);
critical section
flag[i] = FALSE;
remaindersection
} while(TRUE);

Synchronizationhardware
In a uniprocessor multiprogrammed system, mutual exclusion can be obtained by
disabling the interrupts before the process enters its critical section and enabling
them after it has exited the critical section.

Disableinterrupts
Critical section
Enableinterrupts

Once a process is in critical section it cannot be interrupted. This solution

cannot be used in multiprocessor environment. since processes run
independently on different processors.
In multiprocessor systems, Testandset instruction is provided,it completes
execution without interruption. Each process when entering their critical section
must set lock,to prevent other processes from entering their critical sections
simultaneously and must release the lock when exitingtheir critical sections.

do {
acquire
lock
critical
section
release
lock
remainder
section
} while(TRUE);

54
 OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

A process wants to enter critical section and value of lock is false then testandset
returns false and the value of lock becomes true. thus for other processes wantingto
enter their critical sections testandset returns true and the processes do busy
waiting until the process exits critical section and sets the value of lock to false.
• Definition:
booleanTestAndSet(boolean&lock)
{ boolean temp=lock;
Lock=true;
returntemp;
}
AlgorithmforTestAndSet
do{
whiletestandset(&lock)
//donothing
//criticalsection
lock=false
remaindersection
}while(TRUE);

Swapinstructioncanalsobeusedformutualexclusion Definition
Voidswap(boolean&a,boolean&b)
{
booleantemp=a;
a=b;
b=temp;
}
Algorithm
do
{
key=true;
while(key=true)
swap(lock,key);
critical section
lock=false;
remaindersection
}while(1);

55
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

lock is global variable initialized to false.each process has a local variable key. A
process wants to enter critical section,since the value of lock is false and key istrue.

lock=false
key=true
afterswap instruction,
lock=true
key=false

nowkey=falsebecomestrue,processexitsrepeat-until,andenterintocriticalsection.
When process is in critical section (lock=true),so other processes wanting to enter
critical section will have
lock=true
key=true
Hence they will do busy waiting in repeat-until loop until the process exits critical
section and sets the value of lock to false.
Semaphores
Asemaphoreisaninteger variable.semaphoreaccessesonlythroughtwooperations.
1) wait: waitoperationdecrementsthecountby1.
Iftheresultvalueisnegative,theprocessexecutingthewaitoperationis blocked.
2) signaloperation:
Signal operationincrementsby 1,ifthevalueisnotpositivethenoneof the process
blocked in wait operation unblocked.

wait(S){
whileS<=0;// no-
op
S--;
}

signal(S)
{

S++;
}

Inbinarysemaphore count canbe 0 or1. The value ofsemaphore is initialized

to 1.

56
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

do{
wait (mutex);
//CriticalSection
signal (mutex);
//remaindersection

}while(TRUE);
First processthat executeswaitoperationwillbe immediatelygrantedsem.countto 0.
Ifsomeotherprocesswantscriticalsectionandexecuteswait()thenitis
blocked,sincevaluebecomes-1.Iftheprocessexitscriticalsectionitexecutes
signal().sem.count is incremented by 1.blocked process is removed from queue and
added to ready queue.

Problems:
1) Deadlock
Deadlock occurs when multiple processes are blocked.each waiting for a resource
that can only be freed by one of the other blocked processes.
2) Starvation
oneormoreprocessesgetsblocked forever and neverget achancetotaketheir turn in
the critical section.
3) Priority inversion
If low priority process is running ,medium priority processes are waiting for low
priority process,high priority processes are waiting for medium priority
processes.this is called Priority inversion.
The two most common kinds of semaphores are counting semaphores and
binary semaphores. Counting semaphores represent multiple resources,
whilebinarysemaphores, asthename implies, represents two possible states
(generally 0 or 1; locked or unlocked).
Classicproblemsofsynchronization
1) Bounded-bufferproblem
Twoprocessesshareacommon,fixed–size buffer.
Producerputsinformationintothebuffer,consumertakesitout.
The problemarise whenthe producer wantsto put a newitem inthe buffer,but it is
already full. The solution is for the producer has to wait until the consumer has
consumed atleast one buffer. similarly if the consumer wants to remove an item
fromthe buffer and sees that the buffer is empty,it goes to sleep until the producer
puts something in the buffer and wakes it up.

57
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Thestructureoftheproducerprocess
do{
//produceanitemin nextp
wait (empty);
wait(mutex);
//addtheitemtothe buffer
signal (mutex); signal
(full);
}while(TRUE);

Thestructureoftheconsumerprocess
do {
wait
(full);
wait
(mutex);
// remove an item from buffer to
nextc signal (mutex);
signal(empty);
//consumetheiteminnextc
}while(TRUE);

2) Thereaders-writersproblem
Adatabase isto besharedamongseveralconcurrent processes.someprocesses may
want onlyto readthedatabase,some maywant toupdatethedatabase.Iftwo readers
access the shared data simultaneously no problem.if a write,some other process
accessthedatabasesimultaneouslyproblemarised.Writeshaveexcusiveaccessto

58
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

theshareddatabasewhilewritingtothedatabase.Thisproblemisknownas readers- writes

problem.

Firstreaders-writersproblem
No reader be kept waiting unless a writer has already obtained permission
touse the shared resource.
Secondreaders-writesproblem:
Oncewriterisready,thatwriter performs itswriteassoonaspossible.
A process wishing to modify the shared data must request the lock in write mode.
multiple processes are permitted to concurrently acquire a reader-writer lock in
read mode. A reader writer lock inread mode. but only one process may acquire
the lock for writing as exclusive access is required for writers.

Semaphoremutexinitializedto1
o Semaphorewrtinitializedto1
o Integerread countinitializedto0

Thestructureofawriterprocess
do{
wait(wrt);
// writing is
performed
signal (wrt) ;
}while(TRUE);

Thestructureofareaderprocess
do{
wait (mutex) ;
readcount++;
if(readcount==1)
wait (wrt) ;
signal(mutex)
//readingisperformedwait(mutex);
readcount
--;
if(readcount==0)
signal (wrt) ;
signal (mutex) ;
}while(TRUE);
3) DiningPhilosophersproblem

59
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Five philosophers are seated on 5 chairs across a table.Each philosopher has a

platefullofnoodles. Eachphilosopher needsa pairofforkstoeat it.Thereareonly 5
forks available all together. There is only one fork between any two plates of
noodles.
In order to eat, a philosopher lifts two forks, one to his left and the other to hisright.
if he is successful in obtaining two forks, he starts eating after some time, he stops
eating and keeps both the forks down.

What ifall the5philosophersdecidetoeatatthesame time?

All the5philosopherswouldattempttopickuptwoforksatthesametime.So,none of them succeed.

One simple solution is to represent each fork with a semaphore.a philosopher

tries to grab a fork by executing wait() operation on that semaphore.he
releases his forks byexecuting the signal() operation.This solution guarantees
that no two neighbours are eating simultaneously.
Suppose all 5 philosophers become hungry simultaneously and each grabs his left
fork,he will be delayed forever.

ThestructureofPhilosopheri:
do{
wait(chopstick[i] );
wait( chopStick[(i+1) % 5]);
//eat
signal( chopstick[i]);
signal(chopstick[(i+1)%5]);
// think
}while(TRUE);

60
 OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Several remedies:
1) Allowatmost4philosopherstobesittingsimultaneouslyatthetable.
2) Allowaphilosopher topickuphisforkonlyif bothforksare available.
3) Anodd philosopher picks up first his left fork and thenright fork. an evenphilosopher picks up
his right fork and then his left fork.

MONITORS
The disadvantage of semaphore is that it is unstructured construct. Wait and signal operationscan
be scattered in a program and hence debugging becomes difficult.
A monitor isanobject that contains boththe data and proceduresneeded to performallocationof a
shared resource. To accomplish resource allocation using monitors, a process must call a
monitor entry routine. Many processes may want to enter the monitor at the same time. butonly
one process at a time is allowed to enter. Data inside a monitor may be either global to all
routines withinthe monitor(or)localto a specific routine. Monitor datais accessible onlywithin the
monitor. There is no way for processes outside the monitor to access monitor data. This is a form
of information hiding.
If a process calls a monitor entry routine while no other processes are executing inside the
monitor, the process acquires a lock on the monitor and enters it. while a process is in the
monitor, other processes may not enter the monitor to acquire the resource. If a process calls a
monitor entry routine while the other monitor is locked the monitor makes the calling process
wait outside the monitor until the lock on the monitor is released. The process that has the
resource will call a monitor entry routine to release the resource. This routine could free the
resource and wait for another requesting process to arrive monitor entry routine calls signal to
allow one of the waiting processes to enter the monitor and acquire the resource. Monitor gives
high priority to waiting processes than to newly arriving ones.

Structure:
monitormonitor-name
{
//sharedvariabledeclarations
procedure P1 (…) { …. }
procedurePn (…) {……}
Initializationcode(…){…}
}
}
Processescancallproceduresp1,p2,p3……Theycannotaccessthelocalvariablesofthe monitor

61
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Schematicview ofaMonitor

MonitorwithConditionVariables

Monitorprovidesconditionvariablesalongwithtwooperationsonthemi.e.waitandsignal.

wait(condition variable)
signal(conditionvariable)
Every condition variable has an associated queue.A process calling wait on a
particular condition variable is placed into the queue associated with that condition
variable.A process calling signal on a particular condition variable causes a process
waiting on that condition variable to be removed from the queue associated with it.

62
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

SolutiontoProducerconsumerproblemusingmonitors:

monitor
producerconsumer
condition
full,empty;
intcount;
procedureinsert(item)
{
if(count==MAX)
wait(full) ;
insert_item(item);
count=count+1;
if(count==1)
signal(empty);
}
procedureremove()
{
if(count==0)
wait(empty);
remove_item(item);
count=count-1;
if(count==MAX-1)
signal(full);
}
procedureproducer()
{
producerconsumer.insert(item);
}
procedureconsumer()
{
producerconsumer.remove();
}

63
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Solutiontodiningphilosophersproblemusingmonitors

A philosopher may pickup his forks only if both of them are available.A
philosopher can eat only if his two neighbours are not eating.some other
philosopher can delay himself when he is hungry.
Diningphilosophers.Take_forks():acquiresforks,whichmayblocktheprocess.
Eat noodles()
Diningphilosophers.put_forks():releasestheforks.
Resumingprocesseswithin amonitor
Ifseveral processesaresuspendedon condionxandx.signal()isexecutedby someprocess. then
howdowedeterminewhichofthesuspendedprocessesshouldberesumednext?
solution is FCFS(process that has been waiting the longest is resumed first).In
many circumstances, such simple technique is not adequate. alternate solution is to
assign priorities and wake up the process with the highest priority.

Resourceallocationusingmonitor
booleaninuse=false;
conditionavailable;
//conditionvariable

64
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

monitorentryvoidget resource()
{
if(inuse) //isresourceinuse
{
wait(available); waituntilavailableissignaled
}
inuse=true; //indicateresourceisnow inuse
}
monitorentryvoidreturnresource()
{
inuse=false; //indicateresource
is not in use signal(available); //signal a
waiting process to proceed
}

65
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

UNIT-III
Memory Management: Basic concept, Logical and Physical address map, Memory allocation:
Contiguous Memory allocation – Fixed and variable partition–Internal and External fragmentation and
Compaction; Paging: Principle ofoperation – Page allocation – Hardware support for paging, protection
and sharing, Disadvantages of paging.
Virtual Memory: Basics of Virtual Memory – Hardware and control structures – Locality of reference,
Page fault , Working Set , Dirty page/Dirty bit – Demand paging, Page Replacement algorithms:
Optimal, First in First Out (FIFO), Second Chance (SC), Not recently used (NRU) and Least Recently
used (LRU).

LogicalAndPhysicalAddresses

An address generated by the CPU is commonly refereed as Logical Address, whereas the
address seen by the memory unit that is one loaded into the memory address register of the
memory is commonly refereed as the Physical Address. The compile time and load time
address binding generates the identical logical and physical addresses. However, the
executiontimeaddressesbindingschemeresults indiffering logicaland physicaladdresses.

Theset ofall logicaladdressesgenerated byaprogramisknownasLogicalAddressSpace, where

as the set of all physical addresses correspondingtotheselogical addresses is Physical
Address Space. Now, the run time mapping from virtual address to physical address is done
by a hardware device known as Memory Management Unit. Here in the
caseofmappingthebaseregister isknownasrelocationregister. Thevalue intherelocation
registerisaddedtotheaddressgeneratedbyauserprocessatthetimeitissenttomemory
.Let's understand this situation with the help of example: If the base register contains the
value 1000,then an attempt by the user to address location 0 is dynamically relocated to
location 1000,an access to location 346ismapped tolocation1346.
Memory-ManagementUnit(MMU)
Hardwaredevicethatmaps virtualtophysicaladdress

InMMUscheme, thevalue intherelocationregisterisaddedto everyaddressgeneratedbyauser process at

the time it is sent to memory
Theuserprogramdealswithlogicaladdresses;itneverseesthe realphysicaladdresses

66
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Theuserprogramneverseestherealphysicaladdressspace,italwaysdeals withthe Logical

addresses.As we have twodifferent type ofaddresses Logicaladdress intherange
(0tomax)and Physical addressesintherange(RtoR+max)whereRis thevalue of relocation
register. The user generates only logical addresses and thinks thatthe process runs in location
to 0 to max. As it is clear from the above text that user program
suppliesonlylogicaladdresses, these logicaladdresses must bemappedto physicaladdress
before they are used.
BaseandLimitRegisters

Apair ofbaseandlimitregistersdefinethelogicaladdressspace

HARDWAREPROTECTIONWITHBASEANDLIMIT

Binding ofInstructionsand DatatoMemory

Addressbindingofinstructionsand datatomemoryaddressescanhappenat threedifferentstages

Compiletime:Ifmemorylocationknownapriori, absolutecodecanbegenerated;mustrecompile code if

starting location changes
Loadtime:Mustgeneraterelocatablecodeifmemorylocationisnotknownatcompiletime
67
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Executiontime:Bindingdelayeduntilruntimeiftheprocesscanbemovedduringitsexecution
from
one memory segment to another. Need hardware support for address maps (e.g., base and
limitregisters)

MultistepProcessingof aUserProgram

Dynamic Loading
Routine isnotloadeduntilitiscalled
Bettermemory-spaceutilization;unusedroutineisneverloaded
Usefulwhenlargeamountsofcodeareneededto handleinfrequentlyoccurring cases
Nospecialsupport fromtheoperatingsystemisrequiredimplementedthroughprogramdesign

Dynamic Linking
Linkingpostponeduntilexecutiontime
Smallpieceofcode,stub,usedtolocatetheappropriatememory-residentlibraryroutine Stub
replaces itself with the address of the routine, and executes the routine
Operatingsystemneededtocheckifroutineisinprocesses’memoryaddressDynamic linking is
particularly useful for libraries
Systemalsoknownas shared libraries

68
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Swapping
A process can be swapped temporarily out of memory to a backing store, and then brought back into
memoryfor continued executionBacking store –fast disk large enoughto accommodate copiesofall
memory images for all users; must provide direct access to these memory images Roll out, roll in –
swapping variant used for priority-based scheduling algorithms; lower-priorityprocess is swapped out
so higher-priority process can be loaded and executed Major part of swap time is transfer time; total
transfer time is directly proportional to the amount of memory swapped and Modified versions of
swapping are found on many systems (i.e., UNIX, Linux, and Windows)
Systemmaintainsa ready queueofready-to-runprocesseswhich havememoryimageson disk

SchematicView ofSwapping

ContiguousAllocation
Main memoryusuallyinto twopartitions:
Residentoperatingsystem,usuallyheldinlow memorywithinterrupt vector
User processes then held in high memorynRelocation registers used to protect user processes from each
other, and from changing operating-system code and data
Baseregistercontainsvalueofsmallestphysicaladdress
Limit register contains range of logical addresses – each logical address must be less than the limit
register
MMUmapslogicaladdressdynamically
HardwareSupport forRelocationandLimitRegisters

69
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Multiple-partitionallocation
Hole – block of available memory; holes of various size are scattered throughout memory
When a process arrives, it is allocated memory from a hole large enough to accommodate it

Contiguous memory allocation is one of the efficient ways of allocating main memory to
the processes. The memory is divided into two partitions. One for the Operating Systemand
another forthe user processes. Operating Systemis placed in low or high memorydepending
on the interrupt vector placed. In contiguous memory allocation each process is contained ina
single contiguous section of memory.

Memoryprotection

Memoryprotectionisrequiredto protect OperatingSystemfromtheuser processesanduser

processes fromone another. Arelocationregister contains the value ofthe smallest physical
address for example say 100040. The limit register contains the range of logicaladdress for
example say74600. Eachlogicaladdress must be lessthanlimit register. Ifa logicaladdress is
greater thanthe limit register, thenthere is anaddressing error and it is trapped. The limit
register hence offers memory protection.

The MMU, that is, MemoryManagement Unit maps the logicaladdress dynamically, that is
atruntime, byaddingthe logicaladdresstothevalue inrelocationregister. Thisadded value is the
physical memory address which is sent to the memory.

The CPU scheduler selects a process for execution and a dispatcher loads the limit and
relocationregisterswithcorrect values.Theadvantageofrelocationregister isthat it provides an
efficient way to allow the Operating System size to change dynamically.

Memoryallocation

There are two methods namely, multiple partition method and a general fixed partition
method. In multiple partition method, the memory is divided into several fixed size
partitions. One process occupies each partition. This scheme is rarely used nowadays.
Degree of multiprogramming depends on the number of partitions. Degree of
multiprogramming is the number of programs that are in the main memory. The CPU is
never left idle in multiprogramming. This was used by IBM OS/360 called MFT. MFT
stands for Multiprogramming with a Fixed number of Tasks.

GeneralizationoffixedpartitionschemeisusedinMVT.MVTstandsforMultiprogramming with a
Variable number of Tasks. The Operating System keeps track of which parts of memory are
available and which is occupied. This is done with the help of a table that is
maintainedbytheOperatingSystem.Initiallythewholeoftheavailablememoryistreated as

70
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

onelarge blockof memorycalled a hole. The programs that enter asystemare maintained in an
input queue. From the hole, blocks of main memory are allocated to the programs in the
input queue. If the hole is large, then itis split into two, and one half is allocated to the
arriving process and the other half is returned. As and when memory is allocated, a set of
holes in scattered. If holes are adjacent, they can be merged.
Now there comes a general dynamic storage allocation problem. The following are the
solutions to the dynamic storage allocation problem.

 First fit: The first hole that is large enough is allocated. Searching for the holes
starts from the beginning of the set of holes or from where the previous first fit search
ended.

 Best fit: The smallest hole that is big enough to accommodate the incoming
process isallocated. Ifthe available holesare ordered,thenthe searching can bereduced.

 Worstfit:Thelargestoftheavailable holesisallocated.
Example:

Firstandbestfitsdecreasetimeandstorageutilization.Firstfitisgenerallyfaster. Fragmentation
Thedisadvantageofcontiguousmemoryallocationisfragmentation.Therearetwo types of
fragmentation, namely, internal fragmentation and External fragmentation.
Internalfragmentation

When memory is free internally, that is inside a process but it cannot be used, we call that
fragment as internal fragment. For example say a hole of size 18464 bytes is available. Letthe
size of the process be 18462. If the holeis allocated to this process, then twobytes areleft
which is not used. These two bytes which cannot be used forms the internal fragmentation.
The worst part of it is that the overhead to maintain these two bytes is more than two bytes.
Externalfragmentation
All the three dynamic storage allocation methods discussed above suffer external
fragmentation. When the total memory space that is got by adding the scattered holes is
sufficienttosatisfyarequestbutitisnotavailablecontiguously,thenthistypeof

71
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

fragmentationiscalledexternalfragmentation.

The solution to this kind ofexternal fragmentation is compaction. Compaction is a method

by which all free memorythat are scattered are placed together in one large memory block. It
is to be noted that compaction cannot be done if relocation is done at compile time or
assembly time. It is possible only if dynamic relocation is done, that is relocation at
execution time.

One more solution to external fragmentation is to have the logical address space and
physical address space to be non contiguous. Paging and Segmentation are popular non
contiguous allocation methods.
Exampleforinternaland externalfragmentation

Paging
A computer can address more memory than the amount physically installed on the system.
This extra memoryis actuallycalled virtualmemoryand it is a sectionofa hard that's set up to
emulate the computer's RAM. Paging technique plays an important role in implementing
virtual memory.
Paging is a memory management technique in which process address space is broken into
blocks ofthe same size called pages (size ispowerof2, between512 bytes and 8192 bytes).
The size of the process is measured in the number of pages.
Similarly, main memory is divided into small fixed-sized blocksof(physical)memory called
frames and the size of a frame is kept the same as that of a page to have optimum utilization
of the main memory and to avoid external fragmentation.

72
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

PagingHardware

AddressTranslation
Pageaddressiscalledlogicaladdress and representedbypagenumberand the offset.
Frameaddressiscalledphysicaladdressand representedbyaframenumberandthe offset.
LogicalAddress=Pagenumber+pageoffset
PhysicalAddress=Framenumber +pageoffset
A data structure called page map table is used to keep track of the relation between a page
of a process to a frame in physical memory.
Paging ModelofLogicalandPhysicalMemory

73
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Paging Example

32-bytememory and 4-bytepages

FreeFrames

When the system allocates a frame to any page, it translates this logical address into a
physical address and create entry into the page table to be used throughout execution of the
program.
When a process is to be executed, its corresponding pages are loaded into any available
memory frames. Suppose you have a program of 8Kb but your memory can accommodate
only5Kbatagivenpointintime,thenthepagingconceptwillcomeintopicture.Whena

74
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

computer runs out of RAM, the operating system(OS) will move idle or unwanted pages of
memory to secondary memory to free up RAM for other processes and brings them back
when needed by the program.
This process continues during the whole execution of the program where the OS keeps
removing idlepages fromthe mainmemoryand write themonto the secondary memoryand
bring them back when required by the program.
ImplementationofPageTable

Pagetableiskeptinmainmemory
Page-tablebaseregister(PTBR)pointstothepagetable
Page-tablelengthregister(PRLR)indicatessizeofthepagetable
Inthisschemeevery data/instructionaccessrequirestwomemory accesses.Oneforthepagetable and one for
the data/instruction.
The two memoryaccess problem can be solved by the use ofa special fast-lookup hardware
cache called associative memory or translation look-aside buffers (TLBs)
PagingHardwareWith TLB

MemoryProtection
Memoryprotection implementedbyassociating protectionbitwitheach frame
Valid-invalidbitattachedtoeachentryinthepagetable:
“valid”indicatesthattheassociatedpageisintheprocess’logicaladdressspace,andisthusalegal page “invalid”
indicates that the page is not in the process’ logical address space
Valid(v) orInvalid(i)BitInAPageTable

75
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

SharedPages
Shared code
One copy of read-only (reentrant) code shared among processes (i.e., texteditors,
compilers,window systems).
Sharedcodemustappear insamelocationinthelogicaladdressspaceofallprocesses
Privatecode anddata
Eachprocess keeps aseparatecopyofthecodeanddata

Thepagesfortheprivatecodeanddatacanappear anywhereinthelogicaladdressspace
SharedPages Example

76
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

StructureofthePageTable

Hierarchical Paging
Hashed Page Tables
InvertedPageTables

HierarchicalPage Tables

Break up the logical address space into multiple page tables Asimpletechnique
is a two-level page table
Two-LevelPage-TableScheme

Two-LevelPagingExample
Alogicaladdress(on32-bitmachinewith1Kpagesize)isdivided into: a page
number consisting of 22 bits
apageoffsetconsistingof10bits
Sincethepagetableispaged,thepagenumberis furtherdividedinto:
a12-bitpagenumbera10-bitpageoffset
Thus, a logical address is as follows:

wherepiisanindexintotheouterpagetable,andp2isthedisplacementwithinthepageofthe outer page table

Pagenumber page offset

pi p2 d10
12 10

77
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Address-TranslationScheme

Three-levelPaging Scheme

HashedPageTables

Common inaddressspaces>32bits
Thevirtualpagenumberishashedintoapagetable
This page table contains a chain of elements hashing to the same
location Virtual page numbers are compared in this chain searching fora
match
Ifamatchisfound,thecorrespondingphysicalframeisextracted

HashedPageTable
78
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

InvertedPageTable

Oneentryforeachrealpageof memory

Entryconsistsofthevirtualaddressofthepage stored inthat realmemorylocation, withinformation about

the process that owns that page
Decreases memory needed to store each page table, but increases time needed to search the table
when a page reference occurs
Usehashtabletolimitthesearchtoone —or atmostafew—page-tableentries
InvertedPageTableArchitecture

Advantagesand DisadvantagesofPaging
Here isalistofadvantagesanddisadvantagesofpaging−
 Pagingreducesexternalfragmentation, butstillsuffersfrominternalfragmentation.
 Paging is simple to implement and assumed as an efficient memory management
technique.
 Dueto equalsizeofthepagesandframes,swapping becomesveryeasy.
 Page table requires extra memory space, somay not be good for a system having
small RAM.

Segmentation
Memory-managementschemethatsupportsuserviewofmemoryAprogramisa collection of
segments
 Asegmentisalogicalunitsuchas:
 mainprogram
 Procedure
 functionmethod
 object
79
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

 localvariables,globalvariables
 commonblock
 stack
 symboltable
 arrays

User’sViewofaProgram

SegmentationArchitecture
Logicaladdressconsistsofatwotuple:
o<segment-number, offset>,
Segmenttable–mapstwo-
dimensionalphysicala dpd rehsys es ;iecaac hl tambleemntroyrhays:spacebase–
containsthestartingphysicaladdresswherethesegmentsreside inmemory limit –
specifies the length of the segment
Segment-table base register (STBR) points to the segment table’s location in memory
Segment-tablelengthregister(STLR)indicatesnumberofsegmentsusedbyaprogram; segment
number s is legal if s <STLR
Protection
Witheachentryin segmenttableassociate:
validation bit = 0 Þ illegal segment
read/write/execute privileges
Protectionbitsassociatedwithsegments;codesharingoccursatsegmentlevel
Sincesegmentsvaryinlength,memoryallocationisadynamicstorage-allocation problem A
segmentation example is shown in the following diagram

80
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Segmentation Hardware

ExampleofSegmentation

Segmentationwithpaging
81
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Insteadofanactualmemorylocationthesegment informationincludestheaddressofa pagetable

for the segment. Whena programreferences a memorylocationthe offset is translated to a
memory address using the page table. A segment can be extended simply by allocating
another memory page and adding it to the segment's page table.
An implementation of virtual memory on a system using segmentation with paging usually
only moves individual pages back and forth between main memory and secondary storage,
similar to a paged non-segmented system. Pages ofthe segment can be located anywhere in
main memory and need not be contiguous. This usually results in a reduced amount of
input/output between primary and secondary storage and reduced memory fragmentation.

VirtualMemory
Virtual Memory is a space where large programs can store themselves in form of pages
while their execution and only the required pages or portions of processes are loaded intothe
main memory. This technique is useful as large virtual memory is provided for user
programs when a very small physical memory is there.
Inrealscenarios, mostprocessesneverneedalltheirpagesatonce, for followingreasons:
 Error handling code is not needed unless that specific error occurs, some of
whichare quite rare.
 Arrays are oftenover-sized for worst-case scenarios, and only a small fraction ofthe
arrays are actually used in practice.
 Certain featuresofcertainprogramsare rarelyused.

Fig.Diagramshowingvirtualmemorythatislargerthanphysicalmemory.
Virtual memory is commonly implemented by demand paging. It can also be implemented in a
segmentation system. Demand segmentation can also be used to provide virtual memory.

BenefitsofhavingVirtualMemory:
1. Large programs can be written, as virtual space available is huge compared to
physical memory.

82
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

2. LessI/O required,leadsto fasterand easyswappingofprocesses.

3. More physical memory available, as programs are stored on virtual memory, so they
occupy very less space on actual physical memory.

Demand Paging

A demand paging is similar to a paging system with swapping(Fig 5.2). When we want to execute a
process, we swap it into memory. Rather than swapping the entire process into memory.

When a process is to be swapped in, the pager guesses which pages will be used before the process is
swapped out again Instead of swapping in a whole process, the pager brings only those necessary pages
into memory. Thus, it avoidsreading into memory pagesthat will not be used in anyway, decreasing the
swap time and the amount of physical memory needed.

Hardware support is required to distinguish between those pages that are in memory and those pages
that are on the disk using the valid-invalid bit scheme. Where valid and invalid pages can be checked
checking the bit and marking a page will have no effect if the process never attempts toaccess the
pages. While the process executes and accesses pages that are memory resident, execution proceeds
normally.
Fig. Transfer ofapagedmemorytocontinuousdiskspace

Access to a page marked invalid causes a page-fault trap. This trap is the result ofthe operating system's
failure to bring the desired page into memory.

Initiallyonlythosepagesareloadedwhichwill berequired theprocessimmediately.

Thepagesthatarenotmovedintothememoryaremarkedasinvalidinthepagetable.For

83
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

an invalid entry the rest of the table is empty. In case of pages that are loaded in the
memory, they are marked as valid along with the information about where to find the
swapped out page.
When the process requires any of the page that is not loaded into the memory, a page fault
trap is triggered and following steps are followed,
1. The memoryaddress which isrequested bythe process is first checked, to verifythe
request made by the process.
2. Ifits found tobeinvalid, theprocessisterminated.
3. In case the request by the process is valid, a free frame is located, possibly from a
free-frame list, where the required page will be moved.
4. A new operation is scheduled to move the necessary page from disk to the specified
memory location. ( This will usually block the process on an I/O wait, allowing some other
process to use the CPU in the meantime. )
5. When the I/O operation is complete, the process's page table is updated with thenew
frame number, and the invalid bit is changed tovalid.

Fig.Stepsin handlingapagefault

6. The instruction that caused the page fault must now berestarted fromthebeginning. There
are cases when nopages are loadedinto the memory initially,pages are only loaded
whendemandedbytheprocessbygeneratingpagefaults.ThisiscalledPureDemand
Paging.
The only major issue with Demand Paging is, after a new page is loaded, the process starts
executionfromthebeginning. Itisnot abig issue forsmallprograms,but forlargerprograms it
affects performance drastically.

Whatisdirtybit?

84
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

When a bit is modified by the CPU and not written back to the storage, it is called as a dirty
bit. This bit is present in the memory cache or the virtual storage space.
AdvantagesofDemandPaging:
1. Largevirtualmemory.
2. Moreefficientuse ofmemory.
3. Unconstrainedmultiprogramming.There isno limitondegreeof multiprogramming.
DisadvantagesofDemandPaging:
1. Number oftablesandamountofprocessorover head for handlingpage interruptsaregreaterthanin the
case of the simple paged management techniques.
2. duetothelack ofanexplicitconstraintsonajobsaddressspacesize.

PageReplacement
As studied in Demand Paging, only certain pages of a process are loaded initially into the
memory. Thisallowsusto get more number ofprocesses into the memoryat the same time. but
what happens when a process requests for more pages and no free memory is available to
bring them in. Following steps can be taken to deal with this problem :
1. Put the process inthe wait queue, until any other process finishes its execution
thereby freeing frames.
2. Or,removesomeotherprocesscompletelyfromthememorytofreeframes.
3. Or,findsomepagesthat arenotbeingusedright now,movethemto thediskto getfree frames.
This technique is called Page replacement and is most commonly used. We havesome great
algorithms to carry on page replacement efficiently.
PageReplacementAlgorithm
Page replacement algorithms are the techniques using which an Operating System decides
which memory pages to swap out, write to disk when a page of memory needs to be
allocated. Paging happens whenever a page fault occurs and a free page cannot be used for
allocation purpose accounting to reason that pages are not available or the number of free
pages is lower than required pages.
When the page that was selected for replacement and was paged out, is referenced again, it
has to read in from disk, and this requires for I/O completion. This process determines the
qualityofthe pagereplacement algorithm: the lesserthetime waiting forpage-ins,the better is
the algorithm.
A page replacement algorithm looks at the limited information about accessing the pages
provided by hardware, and tries to select which pages should be replaced to minimize the
total number of page misses, while balancing it with the costs of primary storage and
processor time of the algorithm itself. There are many different page replacementalgorithms.
We evaluate an algorithm by running it on a particular string of memory reference and
computing the number of page faults,
ReferenceString
The string of memory references is called reference string. Reference strings are generated
artificiallyor bytracing a givensystemand recording the addressofeach memoryreference.

85
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Thelatterchoice producesalarge numberofdata,where wenotetwothings.

 Foragivenpagesize, weneedtoconsideronlythepagenumber,nottheentireaddress.
 If we have a reference to a page p,thenanyimmediatelyfollowingreferences to
pagepwillnevercauseapage fault.Page pwillbeinmemoryafterthe firstreference;the
immediately following references will not fault.
 Forexample, considerthefollowingsequenceofaddresses−123,215,600,1234,76,96
 Ifpagesizeis100,thenthereferencestringis
1,2,6,12,0,0FirstInFirstOut(FIFO)algorithm
 Oldest pageinmainmemoryistheonewhichwillbeselectedforreplacement.
 Easy toimplement, keep a list, replace pages from the tail and add new pages atthe
head.

OptimalPagealgorithm
 An optimal page-replacement algorithm has the lowest page-fault rate of all
algorithms. An optimal page-replacement algorithm exists, and has been called OPT orMIN.

86
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

 Replace the pagethat will not beused forthe longest period oftime. Use the time
whena page is to be used.

LeastRecentlyUsed(LRU) algorithm
 Pagewhichhasnotbeenusedforthelongesttimeinmainmemoryistheone which will be
selected for replacement.
 Easytoimplement, keepalist,replacepagesbylookingbackintotime.

87
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Secondchancepagereplacementalgorithm
 SecondChancereplacementpolicyiscalledtheClockreplacementpolicy...
 In the Second Chance page replacement policy, the candidate pages for removal are considerin a
round robin matter, and a page that has beenaccessed between consecutive considerationswill not be
replaced.
The page replaced is the one that- considered in a round robin matter - has not been accessed since its
last consideration.
 Implementation:
o Adda"secondchance" bittoeachmemoryframe.
o Each time a memory frame is referenced, set the "second chance" bit to ONE (1)- this will give the
frame a second chance...
o AnewpagereadintoamemoryframehasthesecondchancebitsettoZERO(0)
o Whenyouneedtofindapageforremoval,lookinaroundrobinmannerinthememoryframes:
 IfthesecondchancebitisONE,resetitssecondchancebit(toZERO)and continue.
 IfthesecondchancebitisZERO,replacethepageinthatmemoryframe.
 ThefollowingfigureshowsthebehavioroftheprograminpagingusingtheSecondChancepage replacement
policy:

o Wecanseenotablythatthe badreplacementdecisionmadebyFIFO isnotpresent inSecond chance!!!

o There are a total of9 page readoperations tosatisfy the total of 18 page requests- justas good as the
more computationally expensive LRU method !!!

88
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

NRU (Not Recently Used) Page Replacement Algorithm - This algorithm requires that each pagehave
two additionalstatus bits 'R'and 'M'called reference bit and change bit respectively. The reference bit(R)
is automaticallyset to 1whenever the pageis referenced. The change bit (M) is set to 1 whenever the page
is modified. These bits are stored in the PMT and are updated on every memory reference. When a page
fault occurs, the memory manager inspects all the pages and divides them into 4 classes based on R and
M bits.
 Class1: (0,0)−neither recentlyusednor modified-thebestpagetoreplace.
 Class2:(0,1)−notrecentlyusedbutmodified-thepagewillneedtobewrittenoutbefore replacement.
 Class3:(1,0)−recentlyused butclean-probablywillbeused againsoon.
 Class4:(1,1)−recentlyusedandmodified-probablywillbeusedagain,andwriteoutwillbe needed before
replacing it.
Thisalgorithmremovesa page atrandomfromthe lowestnumbered non-emptyclass.

89
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

UNIT-IV
File Management: Concept of File, Access methods, File types, File operation, Directory structure,File
System structure, Allocation methods (contiguous, linked, indexed), Free-space management (bit
vector, linked list, grouping), directory implementation (linear list, hash table), efficiency and
performance.
I/OHardware:I/Odevices,Devicecontrollers,Direct memoryaccessPrinciplesofI/O
Software:GoalsofInterrupt handlers,Devicedrivers,Device independent I/O software.

File System

File Concept:

Computers can store information on various storage media such as, magnetic disks,
magnetic tapes, optical disks. The physical storage is converted into a logical storageunit
byoperating system. The logicalstorageunit iscalled FILE. Afile isa collectionof similar
records. A record is a collection of related fields that can be treated as a unit by some
application program. A field is some basic element of data. Any individual field contains
a single value. A data base is collection of related data.

Student Marks Marks Fail/Pas

KUMA 85 86 P
LAKSH 93 92 P
DATAFILE

Student name, Marks in sub1, sub2, Fail/Pass is fields. The collection of fields iscalled
a RECORD. RECORD:
LAKSH 93 92 P
Collectionofthese records is calledadatafile.

FILEATTRIBUTES:

1. Name : A file is named for the convenience ofthe user and isreferred by its
name. A name is usually a string of characters.
2. Identifier:Thisuniquetag, usuallyanumber,identifiesthefilewithinthefilesystem.
3. Type: Filesareofsomanytypes.Thetype dependsontheextensionofthefile.

Example:
.exeExecutablefile
.objObjectfile
.srcSourcefile
4. Location : This informationis a pointer to a device and to the location ofthe
file on that device.
90
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

5. Size:Thecurrentsizeofthefile(inbytes,words,blocks).
6. Protection : Access control informationdetermines who cando reading,
writing, executing and so on.
7. Time,Date,Useridentification:Thisinformationmaybekeptfor creation,
last modification,last use.

FILEOPERATIONS

1. Creatingafile:Twostepsareneededtocreateafile.Theyare:
 Checkwhetherthespaceisavailableornot.
 If the space is available then made an entry for the new file in the
directory. The entry includes name of the file, path of the file,etc…
2. Writing a file : To write afile, we have to know2things. One isname ofthe
file and second is the information or datato bewritten on the file, the systemsearches
the entired given location for the file. If the file is found, the systemmust keep a
writepointer to the location inthe file where the next write isto take place.
3. Readingafile:Toreadafile,firstofallwesearchthedirectories forthe file,if the file is
found, the systemneeds to keep a read pointer to the location in the file where the next
read isto take place. Once the read has taken place, the read pointer is updated.
4. Repositioning within a file : The directory is searched for the appropriate
entry and the current file position pointer is repositioned to a given value. This
operation is also called file seek.
5. Deleting a file : To delete a file, first of all search the directory for named
file, then released the file space and erase the directoryentry.
6. Truncating a file : To truncate a file, remove the file contents only but, the
attributes are as itis.

FILETYPES:Thenameofthefilesplitinto2parts.Oneisnameandsecondis Extension. The

file type is depending on extension of the file.

FileType Extension Purpose

Executable .exe Readytorun
.com (or) ready
.bin to run
machine
Sourcecode .c Sourcecodein
.cpp various
.asm languages.
Object .obj Compiled,
.o machine
Batch .bat Commandsto
.sh thecommand
91
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Text .txt Textual

.doc data,
docume
nts
Wordprocessor .doc Variousword
.wp proc
.rtf essor
form
ats
Library .lib Librariesof
.dll routines for
PrintorView .pdf Binaryfile ina
.jpg formatfor
Archive .arc Relatedfiles
.zip groupedintoa
Multimedia .mpeg Binary file
.mp3 containingaudio
.avi oraudio/video

FILESTRUCTURE

File types also can be used to indicate the internal structure of the file. The operating
system requires that an executable file have a specific structure so that it can determine
where in memory to load the file and what the location of the first instruction is. If OS
supports multiple file structures, the resulting size of OS is large. If the OS defines 5
different file structures, it needs to contain the code to support these file structures. All
OSmust support at least onestructurethatofanexecutable filesothatthesystemisable to load
and run programs.

INTERNALFILESTRUCTURE

In UNIX OS, defines all files to be simply stream of bytes. Each byte is individually
addressable by its offset from the beginning or end of the file. In this case, the logical
record size is 1 byte. The file system automatically packs and unpacks bytes into
physical disk blocks, say 512 bytes per block.

The logical record size, physical block size, packing determines how many logicalrecords
are in each physical block. The packing can be done by the user’s application program or
OS. A file may be considered a sequence of blocks. If each block were 512
bytes,afileof1949byteswouldbeallocated4blocks(2048bytes).Thelast99bytes
92
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

would be wasted. It is called internal fragmentation all file systems suffer from
internalfragmentation, the larger the block size, the greater the internal fragmentation.
FILEACCESS METHODS

Filesstoresinformation,thisinformationmustbeaccessedandreadintocomputer memory.
There are so many ways that the information in the file can be accessed.

1. Sequentialfileaccess:

Informationinthe file isprocessed inorderi.e.onerecordaftertheother. Magnetic

tapes are supporting this type of file accessing.
Eg : A file consisting of 100 records, the current position of read/write head is 45 th
record, suppose we want to read the 75 th record then, it access sequentially from 45,
46, 47
……..74,75.Sothe read/writeheadtraverse allthe recordsbetween45to75.

2. Directaccess:

Direct access is also called relative access. Here records can read/write randomlywithout
any order. The direct access method is based on a disk model of a file, because disks
allow random access to any file block.
Eg :Adiskcontaining of256 blocks,the positionofread/write head isat 95 thblock. The
block is to be read or write is 250 th block. Then we can access the 250 th block directly
without any restrictions.

Eg : CD consists of 10 songs, at present we are listening song 3, If we want to listensong

10, we can shift to 10.

3. IndexedSequentialFile access

Themaindisadvantageinthesequentialfileis,ittakesmoretimetoaccessaRecord
.Recordsareorganized insequencebasedonakeyfield. Eg :
A file consisting of60000 records,the master index divide the totalrecords into 6 blocks,
each block consisiting of a pointer to secondary index.The secondary index divide the
10,000recordsinto10indexes.Eachindexconsistingofapointertoitsorginal

93
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

location.Eachrecord intheindex fileconsistingof2 field,Akeyfield andapointerfield.

DIRECTORYSTRUCTURE
Sometimes the file systemconsisting ofmillions offiles,at that situation it is veryhard to
manage the files. To manage these files grouped these files and load one group into
one partition.

Eachpartitioniscalledadirectory.adirectorystructureprovidesamechanismfor organizing
many files in the file system.

OPERATIONONTHEDIRECTORIES:
1. Searchfora file:Searchadirectorystructureforrequiredfile.

2. createafile : Newfilesneedtobecreated,addedtothedirectory.

3. Deleteafile : Whenafileisnolongerneeded,wewanttoremove itfromthe

directory.

4. Listadirectory: Wecanknowthelistoffiles inthedirectory.

5. Renameafile : Wheneverweneedtochangethenameofthe file,wecanchange

thename.
6. Traverse the file system : We need to access every directory and every file
with in a directory structure we can traverse the file system

94
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Thevariousdirectorystructures

1. Singleleveldirectory:

The directory system having only one directory,it consisting of

all files some times it is said to be root directory.

E.g:- Here directorycontaining 4 files (A,B.C,D).the advantageofthe scheme is

its simplicity and the ability to locate files quickly.The problem is different
users may accidentally use the same names for their files.

E.g :- If user 1 creates a files caled sample and then later user 2to creates a file
called sample,thenuser2’s file willoverwrite user1 file.Thatswhy it is not used in
the multi user system.

2. Twoleveldirectory:

Theprobleminsingleleveldirectoryisdifferentusermaybeaccidentallyuse
95
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

thesamenamefortheirfiles.Toavoidthisproblemeachuserneedaprivate directory,

Nameschosenby oneuserdon'tinterferewithnameschosenbyadifferent user.

Root directory is the first level directory.user 1,user2,user3 are user level of
directory A,B,C are files.

3. Treestructured directory:

Two level directory eliminates name conflicts among users but it is notsatisfactory
for users with a large number of files.To avoid this create the sub- directory and
load the same type of files into the sub-directory.so, here each can have as many
directories are needed.

96
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

There are2types ofpath

1. Absoultepath
2. Relativepath
Absoultepath:Beggingwithrootandfollowsapathdowntospecified files giving
directory, directory name on the path.
Relative path:Apathfromcurrentdirectory.

4. Acyclicgraphdirectory

Multiple users are working on a project, the project files can be stored in a
comman sub-directory of the multiple users. This type of directory is called
acyclic graph directory .The common directory will be declared a shared
directory. The graph contain no cycles with shared files, changes made by one
user are made visible to other users.A file maynow have multiple absolute paths.
when shared directory/file isdeleted, all pointers to the directory/ files also to be
removed.

5. Generalgraphdirectory:
Whenweaddlinkstoanexistingtreestructureddirectory,thetree structure is
destroyed, resulting is a simple graph structure.

Advantages:-Traversingiseasy. Easysharingispossible.

97
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Filesystemstructure:
Diskprovides the bulk ofsecondarystorage on which a file system is maintained.
They have 2 characteristics that make them a convenient medium for storing
multiple files.
1. A disk can be rewritten in place. It is possible to read a block from
the disk, modify the block, and write it back intosame place.
2. Adiskcanaccessdirectlyanyblockofinformationitcontains.

ApplicationPrograms

LogicalFileSystem

FileOrganisationModule

BasicFile System

I/OControl

Devices

I/O Control: consists of device drivers and interrupt handlers to transfer

information between the main memory and the disk system. The device driver
writes specific bit patterns to special locations in the I/O controller’s memory to
tell the controller which device location to act on and what actions to take.
The Basic File System needs only to issue commands to the appropriate device
driver to read and write physical blocks on the disk. Each physical block is
identified by its numeric disk address (Eg. Drive 1, cylinder 73, track2, sector
10).

The File Organization Module knows about files and their logical blocks and
physical blocks. By knowing the type of file allocation used and the location of
the file, file organization module can translate logical block address to physical
addresses for the basic file system to transfer. Each file’s logical blocks are
numbered from 0 to n. so, physical blocks containing the data usually do not
match the logical numbers. A translation is needed to locate each block.
98
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

The Logical File System manages all file system structure except the actual data
(contentsof file). It maintains file structure via file control blocks. A file control
block(inode inUnix filesystems)containsinformationaboutthefile,ownership,
permissions, location of the file contents.

FileSystemImplementation:

Overview:

ABoot ControlBlock (pervolume) cancontain informationneeded bythe system to

boot anOS fromthat volume. Ifthe disk does not containanOS,this block can be
empty.

AVolumeControlBlock(per volume)containsvolume(orpartition)details, such as

number of blocks in the partition, size of the blocks, a free block, count andfree
block pointers, free FCB count, FCB pointers.
ATypicalFile ControlBlock

A Directory Structure (per file system) is used to organize the files. A PER-FILE
FCB contains many details about the file.
Afilehasbeencreated; itcanbeusedfor I/O. First, itmustbeopened. Theopen( ) call
passes a file name to the logical file system. The open( ) system call First
searchesthesystemwideopenfiletableto see ifthe file isalready inuse byanother
process. Ifitis,aperprocessopenfiletable entryiscreatedpointingtotheexisting system
wide open file table. If the file is not alreadyopen, the directorystructure is
searchedforthegivenfilename.Oncethe fileisfound,FCBiscopied intoasystem
99
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

wide open file table in memory. This table not only stores the FCB but also tracks
the number of processes that have the file open.
Next, an entry is made in the per – process open file table, with the pointer to the
entry in the system wide open file table and some other fields. These are the fields
includeapointertothecurrent locationinthe file(forthenext read/writeoperation) and
the access mode in which the file is open. The open () call returns a pointer to the
appropriate entry in the per-process file system table. All file operations are
preformed via this pointer. When a process closes the file the per- process table
entry is removed. And the system wide entry open count is decremented. When all
users that have opened the file close it, any updated metadata is copied back to the
disk base directory structure. System wide open file table entry is removed.
System wide open file table contains a copy of the FCB of each open
file, other information. Per process open file table, contains a pointerto
the appropriate entry in the system wide open file
table,other information.

100
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

AllocationMethods–Contiguous
Anallocationmethodreferstohowdiskblocksare allocatedforfiles:
Contiguousallocation–eachfileoccupiessetofcontiguousblocksoBest performance in
mostcases
o Simple– onlystartinglocation(block #)andlength(numberofblocks)arerequired
o Problems include finding space for file, knowing file size,external
fragmentation, need for compaction off-line (downtime) or on-line

Linked
Linkedallocation–eachfilealinkedlist of
blocks o File ends at nil pointer
o Noexternalfragmentation
o Eachblockcontainspointertonextblock
o Nocompaction,externalfragmentation
o Freespacemanagementsystemcalledwhennewblockneeded
o Improveefficiencybyclusteringblocksintogroupsbut
increasesinternal fragmentation
o Reliabilitycan bea problem
o Locating a block can take many I/Os
and disk seeks FAT (File Allocation
Table) variation
o Beginningofvolumehastable,indexedbyblocknumber
o Muchlikealinkedlist,butfasterondiskandcacheable

101
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

File-AllocationTable

Indexedallocation
o Eachfilehasitsownindexblock(s) ofpointerstoitsdatablocks

102
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Free-SpaceManagement
Filesystemmaintainsfree-spacelisttotrackavailable blocks/clusters
Linked list (free list)
o Cannotgetcontiguousspaceeasily
o Nowasteofspace
o Noneedtotraversetheentirelist

1. Bitmap or Bit vector –

ABitmap orBit Vector is series orcollectionofbitswhere eachbit correspondsto adisk block. The bit
cantaketwovalues:0and1: 0indicatesthattheblockisallocated and1indicatesafreeblock. The given instance
of disk blocks on the disk in Figure 1 (where green blocks are allocated) can be represented by a bitmap
of 16 bits as: 0000111000000110.
Advantages–
 Simpletounderstand.
 Finding the first free block is efficient. It requires scanning the words (agroupof8 bits) ina bitmap
for a non-zero word. (A0-valued word has all bits 0). The first free block is then found byscanning for
the first 1 bit in the non-zero word.

LinkedFreeSpaceList onDisk

In this approach, the free disk blocks are linked together i.e. a free block contains a pointer to the
nextfree block. The block number of the very first disk block is stored at a separate location on disk and
isalso cached in memory.
103
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Grouping
Modify linked list to store address ofnext n-1 free blocks in first free block,
plusa pointer to next block that contains free-block-pointers (like this one).
Anadvantageofthis approach is thatthe addresses ofa groupof free disk blocks can
be found easily
Counting
Becausespace isfrequentlycontiguouslyusedand freed,withcontiguous-allocation
allocation, extents, or clustering.
Keepaddressoffirst freeblockandcountoffollowing freeblocks. Freespace list then
has entries containing addresses and counts.

DirectoryImplementation
1. LinearList
Inthisalgorithm,allthefilesinadirectory aremaintainedassinglylinedlist.Eachfilecontainsthe pointers to the
data blocks which are assigned to it and the next file in the directory.
Characteristics
1. Whena new file iscreated,thentheentire list ischeckedwhetherthenew file name ismatchingto a
existing file name or not. In case, it doesn't exist, the file can be created at the beginning or at the end.
Therefore, searching for a unique name is a big concern because traversing the whole list takes time.
2. The list needs to be traversed in case of every operation (creation, deletion, updating, etc) on the
files therefore the systems become inefficient.

2. HashTable
To overcome the drawbacks of singly linked list implementation of directories, there is an alternative
approach that is hash table. This approach suggests to use hash table along with the linked lists.
A key-value pairfor each file in the directory gets generated and stored in the hash table.The key can be
determined by applying the hash function on the file name while the key points to the corresponding file
stored in the directory.
Now, searching becomes efficient due to the fact that now, entire list will not be searched on every
operating. Only hash table entries are checked using the key and if an entry found then the
corresponding file will be fetched using the value.

104
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

EfficiencyandPerformance

Efficiencydependent on:
● Diskallocationanddirectoryalgorithms
● Typesofdatakept infile’sdirectoryentry
Performance
● Diskcache–separatesectionofmain memoryforfrequentlyused blocks
● free-behindandread-ahead–techniquestooptimizesequentialaccess
● improvePCperformancebydedicating sectionofmemoryasvirtual disk,orRAMdisk

I/OHardware:I/Odevices
Input/output devices are the devices that are responsible for the input/output operations in a
computersystem.
Basicallytherearefollowing twotypesofinput/output devices:
 Blockdevices
 Characterdevices
BlockDevices
Ablockdevicestoresinformationinblockwithfixed-sizeandown-address.
Itispossibletoread/writeeachandeveryblockindependentlyincaseofblockdevice.
In case of disk, it is always possible to seek another cylinder and then wait for required block to rotate
under head without mattering where the armcurrently is. Therefore, disk is a block addressable device.
Character Devices
Acharacterdeviceaccepts/deliversastreamofcharacterswithout regardingtoanyblockstructure. Character
device isn't addressable.
Characterdevicedoesn'thaveanyseek operation.
There are too many character devices present in a computer systemsuch as printer, mice, rats, network
interfaces etc. These four are the common character devices.

105
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

DeviceControllers
Device drivers are software modules that can be plugged into an OS to handle a particular device.
Operating System takes help from device drivers to handle all I/O devices.
The Device Controller works like an interface between a device and a device driver. I/O units
(Keyboard, mouse, printer, etc.) typically consist of a mechanical component and an electronic
component where electronic component is called the device controller.
There is always a device controller and a device driver for each device to communicate with the
Operating Systems. A device controller may be able to handle multiple devices. As an interface itsmain
task is to convert serial bit streamto block ofbytes, performerror correction as necessary.
Anydeviceconnectedtothecomputer isconnectedbyaplugandsocket, andthesocketisconnectedto
adevicecontroller.Following isa modelforconnectingtheCPU, memory,controllers,andI/Odevices where
CPU and device controllers all use a common bus for communication.

SynchronousvsasynchronousI/O
 SynchronousI/O−InthisschemeCPUexecutionwaitswhileI/Oproceeds
 AsynchronousI/O−I/OproceedsconcurrentlywithCPUexecution
CommunicationtoI/ODevices
The CPU must have a way to pass information to and from an I/O device. There are three
approachesavailable to communicate with the CPU and Device.
 SpecialInstructionI/O
 Memory-mappedI/O
 Direct memoryaccess(DMA)
SpecialInstructionI/O
This uses CPU instructions that are specifically made for controlling I/O devices. These instructions
typically allow data to be sent to an I/O device or read froman I/O device.
Memory-mappedI/O
When using memory-mapped I/O, the same address space is shared by memory and I/O devices. The
device is connected directly to certain main memory locations so that I/O device can transfer block of
data to/from memory without going through CPU.

106
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

While using memory mapped IO, OS allocates buffer in memory and informs I/O device to use that
buffer to send data to the CPU. I/O device operates asynchronously with CPU, interrupts CPU when
finished.
The advantage to this method is that every instruction which can access memory can be used to
manipulate an I/O device. Memory mapped IO is used for most high-speed I/O devices like disks,
communication interfaces.

DirectMemoryAccess(DMA)
Slowdevices likekeyboardswillgenerateaninterrupttothe mainCPUafter eachbyte istransferred. If afast
devicesuchasadiskgeneratedaninterrupt foreachbyte, theoperatingsystemwouldspend most ofitstime
handlingthese interrupts.So atypicalcomputerusesdirect memoryaccess(DMA) hardware to reduce this
overhead.
Direct Memory Access (DMA) means CPU grants I/O module authority to read from or write to
memory without involvement. DMA module itself controls exchange of data between main memory
and the I/O device. CPU is only involved at the beginning and end of the transfer and interrupted only
after entire block has been transferred.
Direct Memory Access needs a special hardware called DMA controller (DMAC) that manages thedata
transfers and arbitrates access to the system bus. The controllers are programmed with source and
destination pointers (where to read/write the data), counters to track the number of transferred bytes,
and settings, which includes I/O and memorytypes, interrupts and states for the CPU cycles.

107
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

TheoperatingsystemusestheDMAhardwareasfollows−
Step Description

1 Devicedriveris instructedtotransferdiskdatatoabufferaddressX.

2 Devicedrivertheninstructdiskcontrollertotransferdatatobuffer.

3 Diskcontroller startsDMAtransfer.

4 Diskcontroller sendseachbytetoDMAcontroller.

5 DMA controller transfers bytes to buffer,increases the memory address,

decreases the counter C until C becomes zero.

6 WhenCbecomeszero,DMAinterruptsCPUtosignaltransfer completion.

I/Osoftwareisoftenorganizedinthe followinglayers−
 User LevelLibraries −Thisprovides simpleinterfacetotheuserprogram toperform inputand output.
For example, stdio is a library provided by C and C++ programming languages.
 KernelLevelModules−Thisprovidesdevicedrivertointeractwiththedevicecontrollerand device
independent I/O modules used by the device drivers.
 Hardware− This layerincludes actual hardware and hardware controller which interact with
thedevice drivers and makes hardware alive.
A key concept in the design of I/O software is that it should be device independent where it should be
possible to write programs that can access any I/O device without having to specify the device in
advance. For example, aprogramthat readsafile as input shouldbeableto readafileona floppydisk, on a
hard disk, or on a CD-ROM, without having to modifythe program for each different device.

108
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

DeviceDrivers
Device drivers are software modules that can be plugged into an OS to handle a particular device.
Operating System takes help from device drivers to handle all I/O devices. Device drivers encapsulate
device-dependent code and implement a standard interface in such a way that code contains device-
specific register reads/writes. Device driver, is generally written by the device's manufacturer and
delivered along with the device on a CD-ROM.
Adevicedriverperformsthefollowingjobs−
 Toacceptrequestfromthedeviceindependentsoftwareabovetoit.
 Interactwiththedevicecontrollertotakeand giveI/O andperformrequired errorhandling
 Makingsurethattherequestisexecutedsuccessfully
How a device driver handles a request is as follows: Suppose a request comes to read a block N. Ifthe
driver is idle at the time a request arrives, it starts carrying out the request immediately. Otherwise, if
the driver is already busy with some other request, it places the new request in the queue of pending
requests.

Interrupthandlers
An interrupt handler, also known as an interrupt service routine or ISR, is a piece of software or more
specifically a callback functions in an operating system or more specifically in a device driver, whose
execution is triggered by the reception of an interrupt.
When the interrupt happens, the interrupt procedure does whatever it has to in order to handle the
interrupt, updates data structures and wakes up process that was waiting for an interrupt to happen.
Theinterruptmechanismacceptsanaddress─anumberthatselectsaspecificinterrupthandling
routine/function from a small set. In mostarchitecture, this address is an offset stored in a table called
the interrupt vector table. This vector contains the memoryaddresses of specialized interrupt handlers.
Device-IndependentI/OSoftware
The basic functionofthe device-independent software is to performthe I/O functions that are common
toalldevicesandtoprovideauniforminterfacetotheuser-levelsoftware.Thoughitisdifficultto
109
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

writecompletelydeviceindependentsoftwarebutwecanwritesomemoduleswhicharecommon among all the

devices. Following is a list of functions of device-independent I/O Software −
 Uniforminterfacingfordevicedrivers
 Devicenaming-Mnemonicnames mappedtoMajorandMinordevicenumbers
 Deviceprotection
 Providingadevice-independentblock size
 Bufferingbecausedatacomingoffadevicecannotbestoredinfinaldestination.
 Storageallocationonblockdevices
 Allocationandreleasingdedicated devices
 ErrorReporting
User-SpaceI/OSoftware
These are the libraries which provide richer and simplified interface to access the functionality of the
kernel or ultimately interactive with the device drivers. Most of the user-level I/O software consists of
library procedures with some exception like spooling system which is a way of dealing with dedicated
I/O devices in a multiprogramming system.
I/O Libraries (e.g., stdio) are in user-space to provide an interface to the OS resident device-
independent I/O SW. For example putchar(), getchar(), printf() and scanf() are example of user levelI/O
library stdio available in C programming.
KernelI/OSubsystem
KernelI/OSubsystemisresponsibleto provide manyservicesrelated toI/O. Followingaresomeofthe
services provided.
 Scheduling − Kernel schedules a set of I/O requests to determine a goodorder in which to execute
them. Whenanapplication issues a blocking I/O systemcall, the request is placed onthe queue forthat
device. The Kernel I/O scheduler rearranges the order of the queue to improve the overall system
efficiency and the average response time experienced by the applications.
 Buffering −KernelI/OSubsystemmaintainsa memoryareaknownas bufferthat storesdatawhile they
are transferred between two devices or between a device with an application operation. Buffering is
done to cope witha speed mismatch betweenthe producer and consumer ofa data streamorto adapt
between devices that have different data transfer sizes.
 Caching − Kernel maintains cache memory which is region of fast memory that holds copies of
data. Access to the cached copy is more efficient than access to the original.
 Spooling and Device Reservation − A spool is a buffer that holds output for a device, such as a
printer, that cannot accept interleaved data streams. The spooling system copies the queued spool files
to the printer one at a time. In some operating systems, spooling is managed by a system daemon
process. In other operating systems, it is handled by an in kernel thread.
 Error Handling − An operating system that uses protected memory can guard against many kinds
of hardware and application errors.

110
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

UNIT-V
Deadlocks: Definition, Necessary and sufficient conditions for Deadlock, Deadlock Prevention,
Deadlock Avoidance: Banker’s algorithm, Deadlock detection and Recovery.
DiskManagement:Diskstructure,Diskscheduling-FCFS,SSTF,SCAN,C-SCAN,Disk reliability, Disk
formatting, Boot-block, Bad blocks.

DEADLOCKS
Systemmodel:
A system consists of a finite number of resources to be distributed among a number of competing
processes. The resources are partitioned into several types, each consisting of some number of
identical instances. Memoryspace, CPU cycles, files, I/O devices are examples of resource types.
If a system has 2 CPUs, then the resource type CPU has 2 instances.
A process must request a resource before using it and must release the resource after using it. A
process may request as many resources as it requires to carry out its task. The number ofresources
as it requires to carry out its task. The number of resources requested may not exceed the total
number of resources available in the system. A process cannot request 3 printers if the system has
only two.
Aprocess mayutilizearesourceinthefollowingsequence:
(I) REQUEST: The process requests the resource. If the request cannotbe granted immediately
(ifthe resource isbeing used by another process), then therequesting process must wait until it can
acquire theresource.
(II) USE: The process can operateonthe resource.ifthe resource isa printer, the process can
print on theprinter.
(III) RELEASE:Theprocessreleasetheresource.
For each use of a kernel managed by a process the operating system checks that the process has
requested and has been allocated the resource. A system table records whether each resource is
free (or) allocated. For each resource that is allocated, the table also records the process to which
it is allocated. Ifa process requests a resourcethat is currentlyallocated to another process, it can
be added to a queue of processes waiting for this resource.
To illustrateadeadlocked state, consider a systemwith3 CDRWdrives. Eachof3 processes holds
oneofthese CDRWdrives. Ifeachprocess nowrequests another drive, the 3 processes will be ina
deadlocked state. Each is waiting for the event “CDRW is released” which can be caused only by
oneoftheotherwaitingprocesses. This example illustratesadeadlock involvingthe sameresource type.
Deadlocks may also involve different resource types. Consider a system with one printer and one
DVD drive. The process Piis holding the DVD and process Pjis holding the printer. IfPirequests
the printer and Pj requests the DVD drive, a deadlock occurs.
DEADLOCKCHARACTERIZATION:
In a deadlock, processes never finish executing, and system resources are tied up, preventing other
jobs from starting.

111
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

NECESSARYCONDITIONS:
Adeadlocksituationcanariseifthefollowing4conditions holdsimultaneouslyina system:
1. MUTUAL EXCLUSION: Only one process at a time can use the resource. If anotherprocess
requests that resource, the requesting process must be delayed until theresource has
beenreleased.
2. HOLDANDWAIT:Aprocessmustbeholdingatleastoneresourceandwaitingto acquire
additional resources that are currently being held by otherprocesses.
3. NOPREEMPTION:Resourcescannotbepreempted.Aresourcecanbereleasedonly voluntarily
by the process holding it, after that process has completed itstask.
4. CIRCULAR WAIT: A set {P0,P1,…..Pn} of waiting processes must exist such that P0 is
waiting for resource heldbyP1, P1 iswaiting fora resource heldbyP2,……,Pn-1 iswaiting for a
resource held by Pn and Pn is waiting for a resource heldbyP0.
RESOURCEALLOCATIONGRAPH
Deadlocks can be described more precisely in terms of a directed graph called a
systemresourceallocation graph.This graph consists of a set of vertices V and a set of edges E. the
set of vertices V is partitioned into 2 different types of nodes:
P = {P1, P2….Pn}, the set consisting of all the active processes in the system. R= {R1,
R2….Rm}, the set consisting of all resource types in the system.
A directed edge fromprocess Pito resource type Rj is denoted byPi->Rj. It signifies that process Pi
has requested an instance of resource type Rj and is currently waiting for that resource.
A directed edge from resource type Rj to process Pi is denoted by Rj ->Pi, it signifies that
an instance of resource type Rj has been allocated to process Pi.
A directed edge Pi ->Rj is called a requested edge. A directed edge
Rj->Piis called an assignmentedge.
Werepresent each process Pias a circle, each resource type R j as a rectangle. Since resource type
Rj may have more than one instance. We represent each such instance as a dot within the
rectangle. Arequest edgepoints to onlythe rectangle R j. Anassignment edge must also designate
one of the dots in therectangle.
Whenprocess Pirequests aninstance ofresourcetype Rj, a request edgeis inserted inthe resource
allocation graph. When this request can be fulfilled, the request edge is instantaneously
transformed to an assignment edge. When the process no longer needs access to the resource, it
releases the resource, as a result, the assignment edge is deleted.
ThesetsP,R,E:
P={P1, P2, P3}
R={R1,R2,R3,R4}
E={P1->R1,P2 ->R3,R1 ->P2,R2->P2,R2 ->P1,R3->P3}

112
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

OneinstanceofresourcetypeR1
Two instances of resource type R2One
instance of resource type R3 Three
instances of resource type R4
PROCESS STATES:
Process P1 is holding an instance ofresourcetype R2 and is waiting for an instance ofresource
type R1.
ProcessP2isholdinganinstanceofR1andaninstanceofR2andiswaitingforinstanceofR3. Process P3 is
holding an instance of R3.
Ifthe graph contains no cycles, then no process in the system is deadlocked. If
the graph does contain a cycle, then a deadlock may exist.
SupposethatprocessP3requestsaninstanceofresourcetypeR2.Sincenoresourceinstanceis currently
available, a request edge P3 ->R2 is added to the graph.
2 cycles:
P1->R1->P2->R3->P3->R2->P1 P2 -
>R3 ->P3 ->R2 ->P2

113
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

ProcessesP1, P2,P3aredeadlocked. ProcessP2 is waiting forthe resource R3,which isheld by

processP3.processP3iswaiting foreitherprocessP1(or)P2toreleaseresourceR2.Inaddition, process
P1 is waiting for process P2 to release resource R1.

Wealsohaveacycle: P1->R1->P3->R2 ->P1

However there is no deadlock. Process P4 may release its instance of resource type R2. That
resource can then be allocated to P3, breaking the cycle.
DEADLOCKPREVENTION
For a deadlock to occur, each ofthe 4 necessary conditions must held. By ensuring that at least
one of these conditions cannot hold, we can prevent the occurrence of a deadlock.
MutualExclusion–notrequiredforsharableresources;mustholdfornon sharable
resources
Hold and Wait – must guarantee that whenever a process requests a resource,
it does not hold any other resources
o Require process to request and be allocated all its resourcesbefore
it begins execution, or allow process to request resourcesonly when
the process hasnone
o Lowresourceutilization;starvationpossible
NoPreemption–
o Ifa process that is holding some resources requests another resource
that cannot be immediately allocated to it, then all resources currently
being held are released
o Preemptedresourcesareadded tothelist of resourcesforwhich the
process is waiting
o Processwillbe restartedonlywhen itcanregain itsold resources, as
well as the new ones that it isrequesting

114
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Circular Wait – impose a totalordering of all resource types, and require that
each process requests resources in an increasing order of enumeration
Deadlock Avoidance
Requiresthatthesystemhassomeadditionalaprioriinformationavailable
 Simplest andmostusefulmodelrequiresthateachprocessdeclarethemaximumnumber
ofresourcesofeachtype thatitmay need
 Thedeadlock-avoidancealgorithmdynamicallyexaminestheresource-allocation
state to ensure that there can never bea circular-waitcondition
 Resource-allocation state is defined by the number of available and allocated
resources, and the maximum demands of the processes .
SafeState
 When a process requests an available resource, system must decide if
immediate allocation leaves the system in a safe state
System is in safe state if there exists a sequence <P1, P2, …, Pn> of ALL
the processes in the systems such that for each Pi, the resources that Pican
still request can be satisfied by currently available resources + resources
held by all the Pj, with j <I
Thatis:
o IfPiresource needsarenotimmediatelyavailable,thenPicanwaituntilall
Pjhavefinished
o When Pj is finished, Pi can obtain needed resources, execute,
return allocated resources, and terminate
o WhenPiterminates,Pi+1canobtainitsneeded
resources, and so o n If a system is in safe state no deadlocks
If a system is in unsa fe state possibility of deadlock
Avoidance ensurethatasystemwillneverenteranunsafestate
Avoidance algorithms
Singleinstanceofaresourcetype
o Usearesource-allocationgraphMultipleinstancesofaresourcetype
o Usethebanker’salgorithm
Resource-AllocationGraphScheme
Claim edgePiÆRj indicated thatprocessPjmayrequestresource Rj; represented
by a dashed line
Claim edge converts to request edge when a process requests a resource
Request edge converted to an assignment edge when the resource is
allocatedtotheprocessWhenaresourceisreleasedbyaprocess,assignmentedge
reconverts to a claim edge Resources must be claimed a priori in the system

115
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

UnsafeStateInResource-AllocationGraph

Banker’sAlgorithm
Multipleinstances
Eachprocessmustaprioriclaimmaximum use
Whenaprocessrequestsaresourceitmayhaveto wait
When a process gets all its resources it must return them in a finite
amount of time Let n = number of processes, and m = number of
resources types.
Available:Vectoroflengthm.Ifavailable [j]=k,there arek instancesofresource type
Rjavailable
Max:nx mmatrix.IfMax [i,j]= k,thenprocess Pimayrequestatmostk
instancesofresourcetypeRj
Allocation:nxmmatrix.IfAllocation[i,j]=kthenPiiscurrentlyallocated k instances
of Rj
Need:nxmmatrix.IfNeed[i,j]=k,thenPimayneedkmoreinstancesof
Rjtocompleteitstask
Need[i,j] =Max[i,j]–Allocation[i,j]
SafetyAlgorithm
1. LetWorkandFinishbevectorsoflengthmandn, respectively.
Initialize: Work = Available
116
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Finish[i]=falsefori=0,1,…,n-1
2. Findanisuchthatboth:
(a) Finish[i]=false
(b) Needi=Work
Ifnosuchiexists,gotostep4
3. Work = Work
+AllocationiFinish[i] = true
gotostep2
4. IfFinish[i] ==trueforalli,thenthesystemisinasafestate
Resource-RequestAlgorithmforProcessPi
Request=requestvectorforprocessPi.IfRequesti[j]=kthen processPiwants
kinstancesofresourcetypeRj
1. IfRequesti£Needigotostep2.Otherwise,raiseerrorcondition, since
processhas exceeded its maximum claim
2. If Requesti£Available,go tostep 3. Otherwise Pimust wait,since
resources are not available
3. PretendtoallocaterequestedresourcestoPibymodifying thestateasfollows:
Available = Available – Request;
Allocationi= Allocationi + Requesti;
Needi=Needi – Requesti;
o Ifsafe theresourcesareallocatedtoPi
o If unsafe Pimustwait,andtheoldresource-allocationstateisrestored

ExampleofBanker’sAlgorithm(REFERCLASSNOTES)
consider5processesP0throughP4;3resource types:
A (10instances),B (5instances),and C(7instances)

SnapshotattimeT0:
Allocation Max Available
ABC ABC ABC
P0010 753 332
P1200 322
P2302 902
P3211 222
P4002 433
ΣThecontent ofthematrixNeedisdefined tobeMax
–AllocationNeed A
BC
Thesystemisinasafestatesincethesequence<P1,P3,P4,P2,P0>

117
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

satisfiessafetycriteria

P1Request (1,0,2)
CheckthatRequest£Available(thatis,(1,0,2)£ (3,3,2) true

Allocatio Need Available

n
ABC ABC ABC
P0010 743 230
P1302 020
P2302 600
P3211 011
P4002 431
Executingsafetyalgorithmshowsthatsequence<P1,P3,P4,P0,P2>satisfiessafetyrequirement
DeadlockDetection
Allowsystemtoenterdeadlockstate
Detection algorithm
Recoveryscheme
SingleInstanceofEachResourceType
Maintain wait-for graph
Nodes are processes PiÆP
jifPiiswaitingforPj
Periodicallyinvokeanalgorithmthat searches foracycle inthegraph. Ifthere isacycle, there
exists a deadlock
Analgorithmto detect acycle inagraphrequires anorderof n2operations, where n
is the number of vertices in the graph
Resource-AllocationGraphandWait-forGraph

Resource-AllocationGraph Correspondingwait-forgraph

118
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

SeveralInstancesofaResourceType
Available: A vector of length m indicates the number ofavailable resources
of each type. Allocation: An n x m matrix defines the number of resources
of each type currently allocated to each process.
Request:Annxmmatrix indicatesthecurrentrequest ofeachprocess.
IfRequest [i][j]= k,thenprocessPiisrequestingkmoreinstancesofresourcetype.Rj.
DetectionAlgorithm
LetWorkand Finish bevectorsoflength mandn,respectivelyInitialize:
(a) Work=Available
(b) For i = 1,2, …, n, if Allocationiπ 0, then
Finish[i] = false; otherwise, Finish[i] = true
2. Findanindexisuchthatboth:
(a) Finish[i]==false
(b) Requesti£Work
Ifnosuch iexists,gotostep4
3. Work = Work +
AllocationiFinish[i] = true
gotostep2
4. IfFinish[i]==false,for somei, 1£i£n,thenthesystemisindeadlockstate. Moreover,if
Finish[i] ==false,thenPiis deadlocked
RecoveryfromDeadlock:
ProcessTermination
Abortalldeadlocked processes
Abort one process at a time until the deadlock cycle
is eliminated In which order should we choose to
abort?
o Priorityoftheprocess
o Howlongprocesshascomputed, andhowmuchlongertocompletion
o Resourcestheprocesshasused
o Resourcesprocessneedstocomplete
o Howmanyprocesseswill needtobeterminated
o Isprocessinteractiveorbatch?
ResourcePreemption
Selectingavictim–minimizecost
Rollback – return to some safe state, restart process for that
stateStarvation–sameprocessmayalwaysbepickedasvictim, include number
of rollback in cost factor

119
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Secondarystoragestructure:
Overviewofmassstoragestructure

Magnetic disks: Magnetic disks provide the bulk of secondary storage for modern
computer system. Eachdisk platterhasa flat circular shape, like a CD. Commonplatter
diameters range from1.8 to 5.25 inches. The two surfaces of a platter are covered witha
magnetic material. We store information by it magnetically on the platters.

Movinghead diskmechanism

Aread /write head files just above eachsurface ofeveryplatter.The headsare attached to
a disk arm that moves all the heads as a unit. The surface of a platter is logically
divided into circular tracks, which are sub divided into sectors. The set of tracks thatare
at one arm position makes up a cylinder. There may be thousands of concentric
cylinders in a disk drive, and each track may contain hundreds of sectors.

When the disk in use, a driver motor spins it at high speed. Most drivers rotate 60to
200 times per second. Disk speed has 2 parts. The transfer rate is the at which data
flow betweenthe drive and the computer. To read/write, the head must be positioned
at the desired track and at the beginning ofthe desired sector on the track, the time it
takes to position the head at the desired track is called seek time. Once the track is
selected thedisk controller waitsuntildesired sectorreachestheread/write head. The
time it takes to reach the desired sector is called latency time or rotational dealy-
access time. When the desired sector reached the read/write head, then the real data
transferring starts.

120
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

A disk can be removable. Removable magnetic disks consist of one platter, held in a
plastic case to prevent damage while not in the disk drive. Floppy disks are in
expensive removable magnetic disksthat have a soft plastic case containing a flexible
platter. The storage capacity of a floppy disk is 1.44MB.

A disk drive is attached to a computer by a set of wires called an I/O bus. The data
transfer on a bus are carried out by special processors called controllers. The host
controller is the controller at the computer end of the bus. A disk controller is builtinto
each disk drive . to perform i/o operation, the host controller operates the disk drive
hardware to carry out the command. Disk controllers have built in cache, data transfer
atthe disk drive happens b/w cache and disk surface. Datatransfer atthe host, occurs
b/w cache and host controller.

Magnetic Tapes: magnetic tapes was used as an earlysecondarystorage medium. It is

permanent and can hold large amount ofdata. It access time is slow compared to main
memory and magnetic disks. Tapes are mainly used for back up, for storage of
infrequently used information. Typically they store 20GB to 200GB.

Disk Structure: most disks drives are addressed as large one dimensional arrays of
logical blocks. The one dimensional array of logical blocks is mapped onto thesectors
of the disk sequentially. sector 0 is the fist sector of the first track on the outermost
cylinder. The mapping proceeds in order through that track, then through the rest of
the tracks in that cylinder, and then through the rest of the cylinder from outermost to
inner most. As we move fromouterzones to inner zones, the number of sectors per
track decreases. Tracks in outermost zone hold 40% more sectors then innermost
zone. The number of sectors per track has been increasing as disks
technologyimproves, and theouterzoneofa disk usuallyhas severalhundred sectors per
track. Similarly, the number ofcylinders per disk has been increasing; large disks
have tens of thousands of cylinders.

Diskattachment

Computeraccessdiskstorageis2ways.
1. ViaI/Oports(hostattachedstorage)
2. Viaaremotehostinadistributedfilesystem(networkattachedstorage).

1 .Host attached storage : host attached storage are accessed via localI/O ports. The
desktop pc uses an I/O bus architecture called IDE. This architecture supports
maximum of 2 drives per I/O bus. High end work station and servers use SCSI and
FC.

121
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

SCSI is an bus architecture which have large number of conductor’s in a ribbon cable
(50 or 68) scsi protocol supports maximum of 16 drives an bus. Host consists of a
controller card (SCSI Initiator) and upto 15 storage device called SCSI targets.

Fc(fiber channel) is the high speed serial architecture. It operates mostly on optical
fiber (or) over 4 conductor copper cable. It has 2 variants. One is a large switched
fabric having a 24-bit address space. The other is an (FC-AL) arbitrated loop that
can address 126 devices.

Awidevarietyofstoragedevicesaresuitableforuseashostattached.(harddisk,cd
,dvd,tapedevices)

2. Network-attachedstorage:A(NAS)isaccessedremotelyoveradatanetwork
.clients access network attached storage via remote procedure calls. The rpc are
carried via tcp/udp over an ip network-usually the same LAN that carries all data
traffic to theclients.
NAS LAN/WAN CLIENT

NAS CLIENT

NAS provides a convenient way for all the computers on a LAN to share a pool of
storage with the same ease of naming and access enjoyed with local host attached
storage .but it tends to be less efficient and have lower performancethan direct attached
storage.

3. Storage area network: The drawback of network attached storage(NAS) is

storage I/O operations consume bandwidth on the data network. The
communication b/w servers and clients competes for bandwidth with the
communication among servers and storagedevices.

A storage area network(SAN) is a private network using storage protocols connecting servers and
storage units. The power of a SAN is its flexibility. multiple hosts and multiple storage arrays can
attach to the same SAN, and storage can be dynamically allocated to hosts. SANs make it possible
for clusters of server to share the same storage

122
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

DiskScheduling Algorithms

Disk scheduling algorithms are used to allocate the services to the I/O requests on the
disk . Since seeking disk requests is time consuming, disk scheduling algorithms tryto
minimize this latency. If desired disk drive or controller is available, request is served
immediately. If busy, new request for service will be placed in the queue of pending
requests. When one request is completed, the Operating System has to choose which
pending request to service next. The OS relies on the type of algorithm it needs when
dealing and choosing what particular disk request is to be processed next. Theobjective
of using these algorithms is keeping Head movements to the amount as possible. The
less the head to move, the faster the seek time will be. To see how it works,thedifferent
disk schedulingalgorithmswillbediscussedandexamplesarealso provided for better
understanding on these different algorithms.

1. FirstComeFirstServe(FCFS)

It is the simplest form of disk scheduling algorithms. The I/O requests are served or
processes according to their arrival. The request arrives first will be accessed and
served first. Since it follows the order of arrival, it causes the wild swings from the
innermost to the outermost tracks of the disk and vice versa . The farther the location
of the request being serviced by the read/write head from its current location, the
higher the seek time will be.

Example:Giventhefollowingtrackrequestsinthediskqueue,computeforthe Total Head

Movement (THM) of the read/write head :

95,180, 34, 119,11, 123, 62, 64

Consider that the read/write head is positioned at location 50. Prior to this track location
199 was serviced. Show the total head movement for a 200 track disk (0-199).
Solution:

123
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

TotalHeadMovementComputation:(THM)=

(180-50)+ (180-34)+(119-34)+(119-11)+(123-11)+(123-62)+(64-62)=

130+146+85+ 108+ 112+61+2(THM)=644tracks

Assuming a seek rate of5 milliseconds is given, we compute for the seek time
using the formula: Seek Time = THM * Seek rate
=644*5ms
SeekTime=3,220 ms.

2. ShortestSeek TimeFirst(SSTF):

This algorithm is based on the idea that that he R/W head should proceed to the track
that is closest to its current position . The process would continue until all the track
requests are taken care of. Using the same sets of example in FCFS the solution are as
follows:
Solution:

(THM)=(64-50) +(64-11) +(180-11)=

14+53+169(THM)=236tracks

SeekTime=THM* Seek rate

=236*5ms
SeekTime=1,180 ms
Inthisalgorithm, request isservicedaccordingto thenext shortest distance. Startingat 50,
the next shortest distance would be 62 instead of34 since it is only12 tracks away from
62 and 16 tracks away from 34 . The process would continue up to the last track
request. Thereareatotalof236tracksandaseektimeof1,180ms,whichseemstobe
124
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

a better service compared with FCFS which there is a chance that starvation3 would
take place. The reason for this is ifthere were lotsofrequests closed to eachother, the
other requests will never be handled since the distance will always be greater.

3. SCANSchedulingAlgorithm

This algorithm is performed by moving the R/W head back-and-forth to the innermost
and outermost track. As it scans the tracks from end to end, it process all the requests
found in the direction it is headed. This will ensure that all track requests, whether in
the outermost, middle or innermost location, will be traversed by the access arm
therebyfindingalltherequests.Thisisalso knownastheElevatoralgorithm.Usingthe same
sets of example in FCFS the solution are as follows:

Solution:

This algorithm works like an elevator does. In the algorithm example, it scans down
towards the nearest end and when it reached the bottom it scans up servicing the
requests that it did not get going down. If a request comes in after it has been
scanned, it willnot beserviceduntiltheprocesscomesbackdownormoves backup. This
process moved a total of 230 tracks and a seek time of 1,150. This is optimal than the
previous algorithm.

4 .CircularSCAN (C-SCAN)Algorithm

This algorithm is a modified version of the SCAN algorithm. C-SCAN sweeps the
diskfromend-to-end,butassoonitreachesoneoftheendtracksitthenmovesto the
125
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

other end track without servicing any requesting location. As soon as it reaches the
other end track it thenstartsservicing and grantsrequestsheaded to itsdirection. This
algorithm improves the unfair situation of the end tracks against the middle tracks.
UsingthesamesetsofexampleinFCFSthesolutionareas

follows:

Notice that in this example an alpha3 symbol (α) was used to represent the dash line.
This return sweeps is sometimes given a numerical value which is included in the
computation of the THM . As analogy, this can be compared with the carriage return
lever of a typewriter. Once it is pulled to the right most direction, it resets the typing
point to the leftmost margin of the paper . A typist is not supposed to type during the
movement ofthe carriage return lever because the line spacing is being adjusted . The
frequent use ofthis lever consumes time, same with the time consumed when the R/W
head is reset to its starting position.

Assumethatinthisexample,αhasavalueof20ms,thecomputation would be as
follows: (THM) = (50-0) + (199-62) + α
=50+137+20(THM)

=207tracks

SeekTime=THM* Seek rate

=187 *5msSeek Time=935 ms.

The computationofthe seek time excluded the alpha value because it is not anactual
seek or search of a disk request but a reset of the access armto the starting position .

126
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Diskmanagement

Disk formatting: A magnetic disk is a blank slate. It is just a platter of a magnetic

recording material. before a disk can store data , it must be divided into sectors that
the disk controller can read and write. This process is called low level formatting
(or)physical formatting. low level formatting fills the disk witha specialdata structure
for each sector .the Data structure for a sector typically consists of a header, a data
area,atrailer.theheaderandtrailercontaininformation usedby thediskcontroller
,such as a sector number and an error correcting code(ECC). When the controller
writes a sector of data during normal I/O, the ECC is updated with a value calculated
from all thebytes in the data area . when the sector is read ,the ECC is recalculatedand
compared with the stored value. If the stored and calculated numbers aredifferent, this
mismatch indicates thatthe data areaof this sector has become corrupted, and that the
disk sector may be bad. ECC containsenough information, if only fewbits of data have
been corrupted, to enable the controller to identify which bits have changed and
calculate what their correct values should be. The controller automatically does the
ECC processing what ever a sector is read/written for many hard disks, when the disk
controlleris instructedto low level format the disk, it can also be told how many bytes
of dataspace to leave betweenthe header and trailer of all sectors.
Before it can use a disk to hold files , OS still needs to record its own data structures
on the disk. It does in 2 steps. The first step is to partition the disk in to one/more
groups of cylinders. OS can treat each partition as a separate disk. The second step is
logical formatting (or)creation of file system. In this step, OS storesthe initial File
system data structures onto the disk. These data structures include maps of free and
allocate space and initial empty directory.

Boot block:-

When a computer is powered up -it must have an initial program to run. This initial
bootstrap program initializes all aspects of the system, from CPU registers to device
controllers, and the contents ofmain memory, and thenstartsthe OS. To do its job, the
bootstrap program finds the OS kernel on disk, loads that kernel into memory and
jumps to an initial address to begin the OS execution. For most computers, thebootstrap
is stored in ROM. This location is convenient, because ROM needs no initialization and
is at a fixed location that the CPU can start executing when powered up, ROM is read
only, it cannot be infected by computer virus. The problem is that changing this
bootstrap code requires changing the ROM hardware chips. For this reason, most
systems store atiny bootstrap loaderprograminthe boot ROM whose job isto bring ina
fullbootstrapprogramfromdisk. The fullbootstrapprogramisstoredin
thebootblocksatafixedlocationonthedisk.Adiskthathasabootpartitioniscalled

127
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

a boot disk or system disk. The code in the boot ROM instructs the disk controller to
read the boot blocks into memory and then starts executing that code.

Bad blocks:-

A Block in the disk damaged due to the manufacturing defect or virus or physical
damage. This defector block is called Bad block. MS-DOS format command, scans the
diskto find bad blocks. Ifformat findsa bad block, it tellsthe allocationmethods notto use
that block. Chkdsk programsearch for the bad blocks and to lock themaway. Data that
resided on the bad blocks usually are lost. The OS tries to read logical block 87.
ThecontrollercalculatesECCand findsthatthesectorisbad.Itreportsthisfindingto the OS.
The next time the system is rebooted, a special command is run to tell the SCS
controller to replace the bad sector
withaspare.
After that, whenever the systemrequests logical block 87, the request is translated into
the replacement sectors address by the controller.

Sectorslipping:-

Logical block 17 becomes defective and the first available spare follows sector 202.
Then, sector slipping remaps all the sectors from 17 to 202, sector 202 is copied into
the spare, then sector 201 to 202, 200 to 201 and so on. Until sector 18 is copied into
sector 19. Slipping the sectors in this way frees up the space of sector 18.

Swapspacemanagement:-

System that implements swapping may use swap space to hold an entire process
image, including the code and data segments. Paging systems maysimplystore pages
that have been pushed out of main memory. Notethat it may be safer to overestimate
thanto underestimatetheamount ofswap spacerequired, because ifasystemrunsout of
swap space it may be forced to abort processes. Overestimation wastes disk space that
could otherwise be used for files, but it does no other harm. Some systems
recommend the amount to be set aside for swap space. Linux has suggested setting
swap space to double the amount of physical memory. Some OS allow the use of
multiple swap spaces. These swap spaces as put on separate disks so that load placed
on the (I/O) system by paging and swapping can be spread over the systems I/O
devices.

128
OPERATINGSYSTEMSNOTES IIYEAR/I SEM MRCET

Swapspacelocation:-

A Swap space can reside in one of two places. It can be carved out of normal file
system(or) it canbe ina separatediskpartition. Iftheswapspace issimplya large file,
withinthe file system, normalfilesystemmethodsusedto createit,name it,allocateits
space. It is easy to implement but inefficient. External fragmentation can greatly
increase swapping times by forcing multiple seeks during reading/writing of a process
image. Wecanimproveperformance bycachingthe block locationinformationin main
memory and by using special tools to allocate physically contiguous blocks for the
swap file. Alternatively, swap space can be created in a separate raw partition. a
separate swap space storage manager is used to allocate
/deal locate the blocks from the raw partition. this manager uses algorithms optimized
for speed rather than storage efficiency. Internal fragmentation may increase but it is
acceptable because life of data in swap space is shorter than files. since swap space is
reinitialized at boot time, any fragmentation is short lived. the raw partition approach
creates a fixed amount ofswap space during disk partitioning adding more swap space
requires either repartitioning the disk (or) adding another swap space elsewhere.

129