0% found this document useful (0 votes)
53 views83 pagesOperating System
quantum series OS
Uploaded by
sid20037789Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
53 views83 pagesOperating System
quantum series OS
Uploaded by
sid20037789Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 83
B.TECH. (COMPUTER SCIENCE AND ENGINEERING)
FOURTH SEMESTER (DETAILED SYLLABUS)
‘Operating system
‘Course Outcome (CO) Bloom’s Knowledge Level (KL)
At the end of course , the student will be able to understand
€CO.1 _ | Understand the structure and functions of OS KK:
CO2 _ | Learn about Processes, Threads and Scheduling algorithms KK
‘CO3__| Understand the principles of concurrency and Deadlocks K
CO4 | Lear various memory management scheme K
€O5__| Study VO management and File systems KaKe
DETAILED SYLLABUS 3-0-0
Unit Topic Proposed
Lecture
Introduction — Operating system and functions, Classification of Operating systems Batch,
1 _| Interactive, Time sharing, Real Time System, ‘Multiprocessor Systems, Multiuser Systems, | gg
Multiprocess Systems, Multithreaded Systems, Operating System Structure- Layered structure,
System Components, Operating System services, Reentrant Kernels, Monolithic and Microkernel
Systems,
‘Concurrent Processes. Process Concept, Principle of Concurrency, Producer / Consumer Problem,
11 | Mutual Exclusion, Critical Section Probiem, Dekker’s solution, Peterson's solution, Semaphores, | 08
Test and Set operation, Classical Problem in Concurrency- Dining Philosopher Problem, Sleeping
Barber Problem: Inter Process Communication models and Schemes, Process generation.
‘CPU Scheduling: Scheduling Concepts, Performance Criteria, Process States, Process Transition
in| Diagram, Schedulers, Process Control Block (PCB), Process address space, Process identification
information, Threads and their management, Scheduling Algorithms, Multiprocessor Scheduling | 08
Deadlock: System model, Deadlock characterization, Prevention, Avoidance and detection,
Recovery from deadlock
“Memory Management: Basic bare machine, Resident monitor, Multiprogramming with fixed
IV_| partitions, Multiprogramming with variable partitions, Protection schemes, Paging, Segmentation, | 08
Paged segmentation, Virtual memory concepts, Demand paging, Performance of demand paging,
Page replacement algorithms, Thrashing, Cache memory organization, Locality of reference.
VO Management and Disk Scheduling 1/0 devices, and 1/0 subsystems, VO buffering, Disk
V__| storage and disk scheduling, RAID. File System: File concept, File organization and access | 08
mechanism, File directories, and File sharing, File system implementation issues, File system
protection and security
‘Text books:
1. Silberschatz, Galvin and Gagne, “Operating Systems Concepts”, Wiley
2. Sibsankar Halder and Alex A Aravind, “Operating Systems”, Pearson Education
3. Harvey M Dietel, “ An Introduction to Operating System”, Pearson Education
4. DMDhamdhere, “Operating Systems - A Concept based Approach”, 2nd Edition,
5. TMH 5. William Stallings, “Operating Systems: Internals and Design Principles ", 6th Edition, Pearson Education Computer System
Seen ee cee eee c cc naan a ante arena
.. (1-2B to 1-9B)
/* Introduction ; Operating System and Function
* Classification of Operating System : Batch
° Interactive
* Multi-processor Systems
¢ Multi-user Systems
‘© Multi-process Systems
© Multi-threaded Systems
1-2B
A. Concept Outline : Part-1
B. Long and Medium Answer Type Questions .
.... (1-9B to 1-20B)
Part-2 sooo
Operating System Structure
'¢ Layered Structure
'° System Components
© Operating System Services
'« Re-entrant Kernels
Monolithic and Micro-kernel System
A. Concept Outline : Part-2 . 1-9B
1-10B
B. Long and Medium Answer Type Questions
Scanned by CamScanner * An operating system is a program that acts an intermediary
between a user of a computer and the computer hardware.
i Memory management
* ii, Processor management
iii, Device management
iv, File management
© The feature of a batch system is the
the user and the job while that job is executing.
e lack of interaction between
_" Ammulti-processor is a computer system with two or more central
processing units, with each one sharing the common main
nt memory as well as peripherals.
. Ee tis Geet a
is software that manages the computer hardware.
jate mechanisms to ensure the
system and to prevent user programs
Scanned by CamScanner Operating Systems
en
Bootin,
© Uses diagnostic routines to test system for equipment failure,
ii, Copies BIOS (Basic Input System) programs from ROM chips j,
main memory,
iti, Loads operating system into computer's main memory.
2 Formatting :
Formats (initializes) diskettes so they can store data and programs,
computer resource:
Keeps track of locations in main memory where programs and data
are stored (memory management).
Moves data and programs back and forth between main memory
and secondary storage (swapping) via partitioning, using foreground
and background areas, and using buffers and queues.
4. Managing files :
i. Copies files/programs from one disk to another.
ii, Backup files/programs
i. Erases (deletes) files/programs
iv. Renames files.
5. Managing tasks :
May be able to perform multi-tasking, multi-programming, time-sharing
or multi-processing.
QueL2 | Describe the classification of operating system.
OR
Write down the different types of operating system.
Eee
Various types of operating system are as follows :
1. Batch system :
a. This type of operating system was used in the earlier age.
b. To speedup processing, jobs with similar needs were batched
together and were run through the computer as a group.
¢. The definitive feature of a batch system is the lack of interaction
between the user and the job while that job is executing.
d. In this execution environment, the CPU is often idle.
Multi-programming system :
In this type of operating system, more than one program will
reside into main memory.
The operating system picks and begins to execute one of the jobs
in the memory.
Scanned by CamScanner n isa special purpose operating system, used
‘requirements on the ‘operation of a processer
Scanned by CamScanner
Seamed wth camscamer 1
a oat ayera nd ey comma
saeeenantercey hae econ ney
2 i ie ltr. ach pec eles ©
1 aaah ace masteralove elation
TEaTE enn ne tatong terme sa:
1 ponehrating (aera Mae]
fn
Bach of dhe threads ean rum in paral
2
2 ge divide precuring tia not only ements
dao among each thread within aa
‘lock; instead, each processor
fe meeoa poo et tawe,oChat hr opera Pee
Thon inmultithreaded applications, multiple tasks sm PORTE
‘parallel with other applications
FEET] Wet awn the sterence tween marae
‘and multi-programming operating system
‘through various communication
a are runing onthe #7862
7. | Multi-processing refers
Snot multiple| programs in main memory at he
of malls ee time and exces
‘concurrently utilizing single CPU.
27 eter mii CPU, [iin gC
Ttpermits parallel processing. [contest witching takes las
Scanned by CamScanner Fig. 194. Prat sling
Specling is weil tncae device aces ata at diferent ates
‘Te buffer provides awaiting satin where data can ret wile the
lower Soriano. ‘
‘The mart common poing application eprint poting,
‘erases are nad int #
‘Where te brats own ae .
‘Spee a med fo proenng data at remotes. The CPU
‘Mordatvia communion ath ia remote printer
Advantages of spooling over buffering:
{LT spoon pectin wes a disk a very tare bater,
12 Spolingiscapateofvertapping UO operation fr ne job with poser
‘peratne for another a
Scanned by CamScanner
Seamed wth camscamer Sane
‘a way that the Cn
‘Operating system : Refer Q 1.1, Page
Structure of OS:
|L_ Simple structure: MS-DOS written to
1-2, Unit,
provide the most functionality
inte least epace:
‘a. Not divided into modules.
'b. Although MS-DOS has some structure, its
‘functionality are not well separated.
jnterfaces and levels
amen
‘ROM BIOS device drivers
Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer ‘ach program rou an ipa an fer procnting hI
‘Mima sc produces
S a user in
a oe opening a Sle, saving a file
land deleting Be fro the storage disk.
‘bThis i an important task that is also performed by the operating:
system.
4 Communication :
2 Operatig sem os
So Carer mr a aig
®
smay be implemented via shared memory. in which
{oor more proces ad and writs toa shared ection of memory
‘ormessage passing, in which packets of information in predefined
formate are moved between processes by the operating system.
& Error detection :
‘Toe main fanction of operating system isto detect the errors ke
‘bad sectors on hard disk, memory a me
‘Wdevices. ‘overflow and errors related
1b Aer detecting the errors, operating system takes an appropriate
‘ection for consistent computing, |
Scanned by CamScanner
Seamed wth camscamer thy eo ant atte in
fcimcoeomaclnatay ceramics
FEET pitremite between thefallowint:
kernel
rs a a
on
“What ure the differences between shell and kernel?
Scanned by CamScanner
Seamed wth camscamer ‘System call: Refer Q. 1.17, Page 1-178, Unit-
Scanned by CamScanner
Seamed wth camscamer can be done
eaars,| Mode switehing “
| ods atom cal Tisisaapesial
interrupt is over, fe. Process
switching is useful for
‘suspending and resuming of a
process.
Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer ‘TaeERT Piterentinte between te fitowing:
1 Proces and progrem
Lt TATE)
Program aa set of instructions |
‘perform s designated)
"A proce is entirely dependent
Programsare independent. |
‘onprogras.|
\
7 |Preemivanodic tarmentes| Program performs a tak
ey reser Srey ling won
4
7 x petra ent a at of
edna alta
= impale
Scanned by CamScanner
Seamed wth camscamer ‘ted hvn timabuier an the tonsa remit
“The producer cannot dope ada he ufler tf Shery 8
‘onsuier cannot reticte an date the llr tempt
4. the tater ent fl producer can depot a data, "The ensue
‘shouldbe allywed to retrieve data tow four coma
‘Solution to prince :
Readunt cect cas
t
‘count = 0
1
ok
binarvsempbore «= 1:
int producers
Scanned by CamScanner
Seamed wth camscamer L.A ertial section is a code segment in a process in which a shared
tenure is cremed.
2. Each process has u segment of code calle a critieal section. Critical
fection used to avoid race conditionson data items
‘4. Incritiel section the process may be changing common variables,
Updating a table, writing a file and a0 on. At any moment at most ane
proces can execute inertial section (CS).
4. Whenone process s executing in its critieal section, no other process i
to be allowed to execute in its critical section. The execution of erities
section bythe process is mutually exclusive in time.
Axolution to the critical seetion problem must satisfy the followin
three requirements
| Mutual exclusion : If process, is executing ints critical seetion, th
‘no other processes can be executing in their eritical sections.
a Pee tins in is teal eta
Processes wish to enter ther critical sections, then only those proces
{iat are not exesting in their remainder scons ean parted
deciding which will enter its critical section next, and this selec
cannot be postponed indefinitely.
Scanned by CamScanner
Seamed wth camscamer sotto
en ry rns ne ca
[Each of thene graces np indefinitely, repeatedly enter
peeves and Phat when oexacu ecen!uction ot
“eaters the igh and bxamninies the Moekboard.
‘The proceer number ie writen on the lackboard, an that rors
‘Others that proces wil
‘Proceas renters inthe ig to check the Backboard
{repeats thivexeciae until in allowed to oer tactical eeton. Tie
pcre iow ae busy wait
Bury waiting
ating ecure wen a
‘Proviso that inte cater seta!
rode.
_ kind of problem occurs in sultprogramming system,
waitirten:
ceil section
signal imaten),
‘remainder section
1
while TRUE,
|. Busy waiting wastes CPU cyies that some other proces ight be ble
‘touse productively.
Give the principle, mutual exclusion in critical section
problem. Also discuss how well these principles are followed in
‘Dekker's solution,
|
Scanned by CamScanner
Seamed wth camscamer ie sxecution
CONCEPT GUTLINE : PART-2
~ Asemaphore is variable or abstract datatype used 10 control
‘sees to common resource by multiple process in a concurrent
Scanned by CamScanner
Seamed wth camscamer armen?
Seated ified
| Wiat are the disadvantages of semaphores and discuss
‘srultabe technique to overcome them ?
of semaphore:
lit ems tha oe amar Thi
acess ka
ow level. Is ey to make programming i
ust keep track ofall call o wat and to signal the
‘semaphore, Ifthisis nt dae in the correct order,
‘an cause deadlock ‘programmer error
‘Semaphores are used for both condition synchronization and
fxcison ‘These sre distin and ferent events, and i fiat
icapw which meaning any given oo
eo
ae eres i dincteeninge cement
1, Thomuia dinirantage cf thovemephoreisthat
© coca accgeetapeneipiag cease |
ater its critical section must loop continuously in the entry code.
ll
Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer of the lock byte (ether itis 09 1
‘the vale
a
go back to step & ; |
a ee suse of Uherenource ata
Scanned by CamScanner
Seamed wth camscamer Mnlanophera who pend most of eis
yc nt ly ot any ond el a 9H
bees gust sit at the table, pick up the two
loot oa pate then serve and eat Sh
Ma UPCHOPSTICK, CHOPSTICKIG+1 mad 5;
Ear CHOPSTICKL, CHOPSTICREA mel
i. 1 pilepher ay THINK indefinitely. Every philowoie wh
“dz Petephers may PICKUP and PUTDOWN thebr Gases i
easerender, or nendeterminiatically, butthese are atom actions
‘Tue problem is to design a protocol to satisfy the liveneas condita
lany philosopher who tries to EAT, eventually does.
Scanned by CamScanner
Seamed wth camscamer Message passing stem
Bera peng provides & mechasion 1 Oe
_ etn od to roeronie erations without shang tb
Same adress space.
ae ll in tated aie hr
espera process may reside on diferent compara
Hare are several methods for logically implementing a inkand
send (or receive () operations =
{L_ Director indirect communication
‘2, Synchronous or asynchronous communication
3. Automatic or explicit buffering
Scanned by CamScanner
Seamed wth camscamer pmaexfor acess
.
customers)
Aetomerst = customerst —T;
Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer 1 A multilevel queue scheduling algorithm
intoseveral separate queues Fig. 3.3.0,
‘Bach queue has te own ccheduling algorithm,
For example, separate queues might be used
Inackground processe ps
The foreground queue might be scheduled
thelarand quenched an POPS
Scanned by CamScanner
Seamed wth camscamer + Malte Scns
eo ie pave waking ton aa
‘intext ofa CPU in process control block so that a pre
‘be resumed from the same point at a later time.
+2 Anyoperating system that allows for multit
‘ttafcontextevitching te allow diferent processes to
(© Acontext switch occurs when a computer's CPU
‘Processor thread toa different process or thread.
Scanned by CamScanner
Seamed wth camscamer In computer system,
“simultaneously and each
there are various
Scanned by CamScanner
Seamed wth camscamer canto block store the details
(CPU) utiieatin nnd execotion Canufe
to input / output :
lock stores the detail pertaining
and file opened during the process excites
GER neeerive proven aaatrenn sce,
oer |
reenn ness pace means apace that alloted in menor fora
ooo
‘Address paces a epace in computer memory,
A Brery process has an address space.
4. AMaress space can be of two types :
‘a. Physical address space : Physical address space is crested in,
RAM
Ja Virtual address space : Virtual address space san address space
that sereated outside the main memory inside the virtual memory,
anditsereated in hard disk,
1. Process registration involves recording all information necessary to
‘Gentfy a registered process and to differentiate it from other processes
Scanned by CamScanner of thread =
inimize context switching time
iuread provides concurrency within a process,
4 Bien ;
“4 Thismore
“5. The benefits of mult-threading can be greatly
processor architecture.
oe
& Deel
Scanned by CamScanner
Seamed wth camscamer ii
3
ne
@
S
:
2
&
2
%
:
2
22
is
Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer ee
Se ee
| _ Garieen a process to move between the ueuah,
- 5 eae oor mere ei
ta ME
(Fig. 92.1. Multilevel
5 Porexample, each process may start at the top level queue, I'he
ses tucompleted within a given time slice it departs the system,
{¢ Preeses that need more than one time slice may be reassigned the
operating system to a lower priority queue, which gets «lower,
[erentage ofthe processor time !
Scanned by CamScanner
Seamed wth camscamer “hat would be the effect of patting we
SHEN Gave eeu we none avenge od
Siang ohne? (RTS NAS]
mer |
1 Process appears twice in the ready queve and is scheduled twice a
‘often an ater preceses.
2 Ineffect, that process will have increased its priority since by wetting
{ewe mare offen itis receiving preferential treatment 1 wail druble tbe
time given to that process.
|4, Treadvantage of this acheme a that more important obs could be ven
‘hore tne Iewll provide higher peority with minimal modification te
Sthedaler.
14. Trediadvantage ofthis schemes thatthe shorter job wil muffler bene
Aaarant jobs il execute with mare time. There willbe overhead for
aiaaging pointers Itmay also increase overhead if same process Fane
ack torback.
{BBL Consider the set of the processes given in the table and
the following scheduling algorithms :
i" Round Robin (Quantum = 1)
fi, Round Robin (Quantum =2)
li, Shortest Remaining Job First
eels te
‘Process 14 | Arrival Time | Execution Time
0
2
3
3s
4
mooe>.
Irthereis tie within the processes, the tie is broken in the favour of
the sldest process, Draw the Gantt chart and find the average
Waiting time, response time and turnaround time for the
algorithms. Comment
“on your result, Which one is better and why ?
‘Round Robin (Quantum = 1)
Prep Ta
oraseseTs et
Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer 4 Circular wait:Aset IP, Py.-P, of waiting processes mustexst veh
‘hat P, is waiting fora cexource Bld by P,P, i waiting fora resource
‘bold by oP is waiting for a resource bel by P, and Ps waiting
fora resource held by Py.
GEER] How deadlock are prevented?
on
etn and explain th dea prevention.
What are the approuches that can be used for prevention of
B+9+2+6 +5556,
desk? (ARTUR)
oresacet Bese
What are the necessary, Deadlock prevention isan approach which ensures that sytem wil ever
‘enter indeadlock state.
‘Thew are following approaches of deadicee prevention:
41 Mutual exclusion :
‘2 Mutual exclusion condition must old fr non-sharable resources.
1 access toa resource requires mutual exclusion, then mutual
‘exclusion must be supported by te operating system.
2 Hold and wait:
‘The hold and wait condition canbe eliminated by forcing a process
toreleaseall resources held by it, whenever it requests a re80ur3©
‘hat isnot arable
Forexample, process copies data from afopp disk to hard disk,
sort disk le and then prints the reelts toa printer.
‘all the resources must be requested at the beginning of the
process, then the process must initially request the floppy disk,
harddisk anda printer.
1 wil bold the priate rite entire execution, even though itneeds
the printer only atthe end.
& Nopre-emption:
‘Ifa process holding certain resources i denied a further request.
‘Thal process must release it orignal resources and if necessary
‘request them again, together with aditional resource.
fa process requests a resource that is currently hed by another
proces, the operating stem may preempt the secund process and
equire it to releaso its resources.
‘& Preemption is posible for certain types of resources, such as CPU
and min memory.
4 Circular wait: -
| One ayo prevent the circular wait condition ish Hinear ordering
of differen types of ester resources,
Scanned by CamScanner
Seamed wth camscamer Dn Resareloation gra x und a descr he dean alin
{hick scm eure allocation grag
‘2 Graph consi owt oCeertcn ¥) and net of wes.
3 the etive recuse tent nated by PP, Py Pad
‘me ansaoalresore spe the tended by
44 Holding of erarc by proces ane y aaignment we
Reusing of rescue by proces i deca by request sae
1k Porrepresnting pron nd reearee inthe rsnurcalestionerath
‘ito ty equate ad cr Bach preeaetveprenemt iy ich and
‘cscrce ty mare Do within the hare reprewots the mamber of
Conner thefatowing system which consist ree proces
Bund P, and four remurceste Ry Ry Rand, Henures Rao
‘Nive on nntance, haste instance and Rh three rss
1 process Pisin stance f resource type sn Saweiting
Taran nance a reaoures type
Process, heing a intance S¢, and, and waiting for an
festancs a resare pe,
+P Ry Ppp + Py RAPD
Bs
®
Scanned by CamScanner
Seamed wth camscamer ® o
4, The ascumptions made inthe wait for graph areas alloms =
a Anedg from PtoP na wait fr graphimplies that process P,
{s waiting for process? to release a resource that P, needs.
1b Anedge P+ exits i a wait-for graph if and only if the
‘eeoure allocation graph contains two es
FOE di +E focmen rere
6. irwrdcnlde pete hegoen bs endkedahervta
ae
AL. Several instances of resouree type:
11 Unmarkallactive processes from Allocation, Max and Available in
‘accordance with the system state,
Scanned by CamScanner
Seamed wth camscamer For’ =0,1, 1, if Allocation, 0, then Finish) = fale:
thera, Finish = trae,
Find an index uch that both
Fnlshlil == fale
aquest = Work
aasuch exists, goto step)
Work = Work + Aloation,
Finish = troe
Goto sep (2).
Finish == alse for some 04
[Need > Available-+ 381 ¢ a10
“Hence, resource will not be granted,
bP <21,0>
[Need < Available -+ Tru
priority processes et
blocked, and high priority |
process
‘Baforcement of priorities
‘uncontrolled resource’
Scanned by CamScanner
Seamed wth camscamer ‘Explain performance criteria,
Refer Q. 3.1 of CEU eehedeting
tig: What lnCPU scheduling? Write the aiference
Cmptive and non pre-emptive: eneenet
aus Refer Q 3.1 and Q. 35. eae
Explain process transition diagram
BS eter. 35.
tht, Discuss Process Control Black (PER.
fase Befer 37 "
{Q.5. Write short notes on process identification information.
Se eer a. 5:10.
6. Write short notes on fllowing
§; FOVS scheduling
i, SUF scheduling
i Round robin scheduling
a
H uefer@. 817. H.R. S18. Hh Refer. 320
G1. Write the difference between following:
1. Thread and process
1 Dice level thread and keenct level thread
2 ee form, short term and mid term scheduler
at
ane
Ker Q.31s. 2 Refer Qars, Refer 3s
6 Baplnin Hk’ slortbn fr evokes of Geen
Scanned by CamScanner
Seamed wth camscamer “Memory management is mainly concerned withthe allocation of main
Imemory to reqocsting proces, No process can ever run, before &
{rtain amount of memory is allocated toit.
‘ho overall resource utilization and other perfarmance criteria of
‘ampater system are largely aflected by te performance of the memary
‘management module
Five requirements of memory management are:
1. Relocation :
‘Programmers do not know where the program wil be placed in
‘memory when its executed.
Wile the program executing, t may be rwapped te dik and
abs returned to main memory at diferent location (relocated)
| & Mery rfrence mathe tanated inte ade aan,
Scanned by CamScanner
Seamed wth camscamer ‘Address binding ie part of computer m
to the opera = memory management andi ie
aarees to acne. Teno allt te apltioon tak,
Instructions and data to
a aay ‘memory addresses can be done in following
i. Compile time : When itis known st compile time where the
rs will reside, comy
ere pile time binding iv ved to generate the
sh Load time + When it eno kamen at empl tine hare
preces il ee n sraoy thn the compl roan
Si Execution time Ifthe proces cane moved during taxation
frmene menorysqguet omaha ing mt bape
to be done at run time. =
Write a short note on :
4. Dynamic loading
i, Dynamic linking
A Dynamic loading :
4. Indynamic loading, a routine of «program s not loaded until itis
called bythe program.
[Allzoutines are keptondiskin a relocatable load format. The main
‘program is loaded into memory and is executed.
Other routines methods or modules are loaded on request,
‘Dynamicloading makes better memory space utilization and unused,
routines are never loaded.
‘It is useful when large amounts of code are needed to handle
infrequently occurring cases.
No special support from the operating system is required,
‘implemented through program design.
©
a oe ee
Scanned by CamScanner
Seamed wth camscamer 4. Dynamic inking:
2 Linking isthe
proces of ecleting and combining var
‘ode and data now exeewable eth canbe lade,
and executed.
Operating system can link aystem level libraries to p
Wihan it combines the irae at ond ie the Linking ing
inking and when tis linking is done athe ime of
itis called as dynamic linking. aig
4. Instatic linking ibearies linked a compile ime. so
‘ie becomes bigger herearin dynamic king rare egeat
‘execution time so program ende size remains smaller Ne
WEETAT] Write te short note on
ee
1 Overlays is used to keep in memory only those instructions ant
data that are needed at any given time
2. When other instructions are needed, they are loaded into spay
‘that was occupied previously by instructions that are no longer
needed.
‘8. Overlays are implemented by user no special support needed rae
operating system, programming design of overlay structure i,
ik Swapping:
1A process must be in memory to be exceuted
% Aprocess, however, can be swapped temporarily out of memoryto
‘backing store and then brought back into memory for continved
execution.
Swapping makesit posible forthe total phyical addres pe
ail process to exced the real physical memory of the Yate
{hosinereasiag the degre of maltprogramming in system
Seretimos wap out proces aio known ws roll-out and maps
process known as ral in "
anicstT Sem)
eo 88M
Pryncal addres spac ina set fal pyri sess corresponding’
2 Tpeselogeal addreses
eal adress wenerated by CPU: Physica adress ithe arn of
3 bagepemiory and tis loaded into the memory address existe
Compile time and load time ness binding methods generat ame
4 (fea and physical adresses.
on tine adress binding scheme results in different logical and
5 ipsa addresses.
mverally local address referred asa vrtal adres
der alogealudress pee of ght page of 10%
sich mapped ote a pysiea memory of3 frames then
deo role ane nga sare?
a How many bits are in physical address ?
5 (a a a)
Taewer
PP trelogcal address pitino two pars: the page address and then the
ee
‘re page ares mst beable reference or distin pen Tis
ures adress its vor
efit mtb abl toreference 1,024 (or 2" distinct word ee
8 ett equres 10 adres te Therefore 8bis are reared or
Theta ares =
Toe pial reais alo spt nto two parts: the fame adres
then te oe
The munher frames 92(or 0 tha wil require Sits,
same asthe sizeof he page therefore thie will
4 Thesizeofthe frame
require 10bits.
1. Therefore, 15 bits are required forthe physical adres.
WET] erine u
i. Bare machine
fi, Resident monitor
Tarver]
{Bare machine +
1. ‘Thisis the simplest form of memory management.
2 Thused by hardware diagnostic, by system boot code, real time!
edicated systems.
terms:
Scanned by CamScanner Placement algorithm with partitions:
‘a. Bqualsize partitions:
1 tial thera abe pein pesam
4 Because all partitions are equal siz, oes not matter which
partion sued.
* Sooetunreate ieee
|b Unequal-size partitions, use of multiple queues:
‘Tisalgorthm assign each proces tothe sale patton within
vehi wil
|i Aqueue exist for each partition se
4% etsen to minimize internal fagmentation.
ji, Problem with this algorithms that some queues mht be empty
‘while sore mig be oad.
“6 Unequal-size partitions, use of a single queue:
|L-Inthisalgorithm, when etme to lad process into memory, the
sullest avilable partition tha wil old the process elected
AL Teineeassthe evel of ultiprogramming atthe expense of steal
fragmentation.
Scanned by CamScanner
Seamed wth camscamer 2 meee, Nagel comorioe eto es taro sae
3 Weriiears Teoma moyen
4 octamer nrarneatntae
6 Wisplecdie ihe than Memeo ee
poe ae
ante remy nd eter ts nevi dd cin
Tear, th te Semele etiesoag ree
FEE vice yume memory alan eng
on
‘Discuss first fit, best fit and worstfit memory allocation techniques.
1 Prat ft, best it, and worst tae the most common strategies used to
Select a fee bole fom the set of avaiable oles.
2 Fast fit and best fit are better than worst Bt in terms of speed and.
‘sorage utilization respectively.
Dynamic memory allocation techniques
1. First ft: Allocate the first ole that isbig enough. Searching canstart
titer at the beginning of the st of holes oa the leaton where the
first t search ended. We can stp searching soon. aswe ind
‘afree hole that is large enough,
2 Best fit: Allocate the smallest hole that isbig enough. We must search
“the entire list, unless the list is ordered by sie. This strategy produces
‘the emallest leftover hole.
| Worst fit: Allocate the largest hole. Again, we must search the entire
reget el
= me td eT crore eos
Scanned by CamScanner
Seamed wth camscamer Process P¢has no memory}
‘Twomemory partition of size 100 K and 300K are fre which ant be
‘Tlocrted to Pl because it has ice more than these memory partitions.
Best fit:
Scanned by CamScanner
Seamed wth camscamer ies = |
——— ——
oer 7 ean index into the pag tte and isthe dolce thin
erore
‘What is paging ? Describe how logieal address ts
Irasated to physical adress in a paged system. Further
Hamutovty prceemeaiveniplasoreras on
-Q. 4.16, Page 4-198, Unit-d.
‘of logical address to physical address :
Scanned by CamScanner
Seamed wth camscamer ee
Pipette diene :
ie erating yer erie the memary cen tering Pa,
fan
rid free frame in oeated and VO is requested
8 Wit Noa ito the fre frame, 2 ae
jon af VO, the process table and page table are pated
Veet instructinse restarted. a
> a page fault by operating system +
Mechs neo able asl ep ith pee ate
Wechecrocee tn Setcrmine whether the reference is vali oF 3%
reall memory access.
+
“
page iron
® backing store
Scanned by CamScanner
Seamed wth camscamer iS —___seo terest
‘TLB contains only fw ofthe page-tabe entries. When a
yearns cea eee na
a
number i found Oxnown as TB i) its frame numbe a
4, 10h [oth avalabe and is weds ence emery =
swe modify the internal table kn ‘page number is notin the TLE (known asa TLB miss) a memory
toindiate thatthe pravis now ingest 30. WIMP. the pe table must be made
th wa iterated by tee. The ris te frase number i oMaied, we can we to aces memory.
page as though ited always been in merger 1h Jpadaiton we add the page number and frame number to he TLB, so
1. Iasi be nnd guy on th net fren
sa TLB i lendy fl of entre, the operating apse mast mac
Meter rel
septacerent policies range from least recently used (LRU) to random.
in the TLB consists oftwo parts: key (r tag) and vlog
associative memory is presented with an iter, th item
keys simultaneously.
Jaswer
me ofthe most common structures of page tables are
Hierarchical page table :
%
the corresponding value field is returned. 4 en pnd hierarchical page table the logical adress space is divided
the number of entries in « TLB is small, often numbering multiple page tables
164 and 1,024.
Je technique used is a two-level page table
used with page tables inthe following way ¢._ Tworlevel paging as shown in Fig. 4.20.1
age tle
Fig. 4.20.1. Two-level paging scheme
4 Alogical address (on 32-bit machine with 4K page size) in tort
isdivided into following two parts
Scanned by CamScanner ~ sn en ee ae
prom pa naa
When a memory reer ca pa el dre
triton put Sapo psoas pes
Seer tent hen aero
Af mach Gund eyo ary en pli nd
= is generated. If no match if found, error is indicated.
‘BEGET wat doy mean segmentation ivetbehardare
Inpeentation of omen
Scanned by CamScanner
Seamed wth camscamer adres space within
‘Bach segment is treated as single, smal linea
rn dese space is
“the container logical adress space proper, and tis
>
4 Medico ramon the minor pes ne Sewing
scheme.
Scanned by CamScanner
Seamed wth camscamer ‘aleatate,
adres
‘OS maintain a list of free holes ia :
=
nnn nem tee
ins ain
Sei aoe hee ase ae
Saya ttre eg nice nent
Bee on ae cee
es
i
pena per
Orr viru aasrets
i. Segment number: Virtua ares space consist of apt 16 segments
, 16-2
{bits are needed to specify segment number.
fi Page number: Hardware pages each seyment into S\2bytes pages
512 = 2 byte pages
‘Sire of segment is 2"* bytes.
Ff 2/2 = 64 "= T pages
“bits are required to specify the page number.
{M, Offeet within page: For byte page, hits are needed,
I tne vl nro nomi oe mae Ai
44749220 |
Wi aresementtonandpaing omnes raet
alll
Scanned by CamScanner
Seamed wth camscamer aoa ogres the
ee ardor o al gael ers oe oe
eS ccektres taal eee
vial wena em Rees «eee ea
pie nl carn Serra ea
done dynamically, wil programsarebeing xen the CPU.
of virtual memory:
“Alors sto it many large programs ito aelatively small RAM,
4 Only part of «program needs tobe loaded into memory.
Memory can be used efficient because a setion of program eae
aly when it is needed in CPU,
44 Virtual memory allows sharing ofcode and data, unlimited amounts
mult: programming: 7
TAG] Explain the process of demand pain, «
‘The process of loading the page into memory on demand (whenever Pas
ess m= eo a "The process inciodes the
‘memory, it generates an interrupt indicating memory access alt,
2. TheOS puts the interrupted process in. blocking state. Por the execs}
toprceed the OS must bring the required page ita Ube memory
Scanned by CamScanner
Seamed wth camscamer re ee
a
iment nop ft en
mene
[Consider a demand poged system. Page tales are held
gute. 1¢ takes 8 milliseconds to service page fault fan
im bay pace is available or the replaced page is not modified, and’
cart condsif the replaced pageis modified. Memory access ines
aignoneconds. Assume thatthe page to he replaced ismoified 70
une of te ine, Whats he maximum acuptale page fasts
pereretfective access time of not more than 200 nanoseconds?
Fee pve conditions, we et
D2 ye=(1—P) x 0.1 ys (0.90) x8 millisee + (0.7P) x20 millisee
T= =p + 2400 P + 14000
F100 ns-=0.1 psec and 200 ns = 02 psec]
1 = 16390
p= 0.00006
GeGA, | Consider a demand paged system. Page ablesare stored
Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer [tring for f0¥r page frames using
A BT TRNTB IIA 42
2
r
fa]
7
[2]
ber of page fault will be 13.
ae a eee
a 5 3) I)
2 z a
3 3] Palle
4 HC]
542
a
z
3
«
‘otal umber of page fault will be 13
Consider the following reference string 1234215621
all
Scanned by CamScanner
Seamed wth camscamer What do you mean
Cbeate tener
Differentiate betm
OS fer 428, Paging and,
ioerrs dered
arenccaee ss me Maes
Br)
pox FeforQ t27 and 48, stvanoge and
Discuss the
Se eter 5:2, 2” PMSMeRE alpine,
@.2. What is thrashing ? State the
solution. State the cause of a
pos, Refer Q. 4.57, ae
200
Scanned by CamScanner Ee
ts ict tia
Kee mes ran oe
Eee
rear
ios chetoc ga
Pye rcrsehetuleg
STF scheduling
SCAN scheduling
Explain the terms VO devices. What are the c
aodevices?
1 lapuvoutput, or VO, is the collection of interfaces that different
functional urits (sub-systems) ofan information processing system
‘eta communicate with each other, or the sguals information sent
through those interfaces,
1b Imptsare the signals received by the unt, and output are the signals
sot from it
“1 Thetorm can also be used as part of an action; todo VO" iso perform
__atinput or output operation
TO devices are used by « person (or other system) to communicate
‘witha computer.
Scanned by CamScanner ‘Scanned by CamScanner
Seamed wth camscamer ‘overvrite buffers before they bad been consumed.
- ig The producer can ony fill up tober /—1 while data buffer)
jewaiting t be
Scanned by CamScanner
Seamed wth camscamer pee simplest frm fdisk scheduling isthe Fist. Come Fis Served
Feces algorithm.
iqueve 96, 183,57, 122,14, 196,65, 67
starts at 5
ee im
‘Fig. 5.6.1. FCPS disk seh
‘This algorithm is intrinsically fair, but it generally does not provide :
the fastest service. aa
Consider, for example, a disk queue with requets for UOtablocks
em 1 1d, a 12, 65,67 that ode.
frst move from S300
Scanned by CamScanner
Seamed wth camscamer quove = 98, 183,37, 12,14,
head starts at 53 el
9 14 87 536567 98 tania. ans
OK scheduling
Inthis the disk arm reverses direction
all the way to the end ofthe disk.
and C-LOOK scheduling, beeause the
_ delore continuing to move ina given direction.
Scanned by CamScanner
Seamed wth camscamer ‘Come, First Serve (FCFS)
‘286, 147, 91,177, 94,150
Scanned by CamScanner
Seamed wth camscamer ieee (66 - 62) + (165. v #
(120-1) yuan tes 8 #89 a
<4 900110 ees
O22 2 65 ne ag.
‘Total head movement _
= (68-62) + (65-92) + (92-12) ¢<490-19)
+ (142 ~ 120) +485 —149)
= 4344204108226 %=21
Assume that head initially moves towands zo,
0 12 32 62 66 120 12 165
‘Fig. 5.103, =|
t |
nna _ asa |
+ (120 ~ 66) + (142—120) +
= 30+ 20+ 12 + 66454422:
=m
Scanned by CamScanner
Seamed wth camscamer se, blocks 2,3, 45,8, 9,10,
“Rare fre and thereat ofthe blocks were
ats sitoation, we would keep aprinter toc:
2 would contain a pointer ablackd ehh
Boiron pin works tick ls ae
‘is scheme is not efficent; to traverse the lt,
* eck, which requires substantial VO time
the bit is0,
naam
Scanned by CamScanner
Seamed wth camscamer orm of RAID datas striped
Ne te
sreater than ach data word it having its ova hemig dea
te than cote vere Bobi
ecg, | * beta the data urracyend dre
“« ‘ths ret te tre er ia saa
e ertarive failures 7
|g Minimum two drives are required inthis 7
aa po 2
rr
: the performance J rye data block i epit and in writn onda dk,
E «Stipe party's generated on writes and thi wring iene
sierra | «See ie pert mt
ty and higher datz, | ARAIDSarray can recover fom har rv file miele
higher inenvironments where applications need igh
now stands on nae
“Inthis level the read/write speeds igh.
dozen RAID level schemes IDLevel 4:
prevailing RAID lee Inthis RAID level the requirement of three drives ia seenandin
this evel, the entire block i written ona disk,
"Party is generated in writes and recorded on parity disks and
-hecked on reads. :
This level of RAID has high reading spends ands bil fice
52
ei adem
rit data disks and parity’ -don reads.
_Ghtrbted onto three dives edi :
Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer fea San ith
fe Sn vs
poral een
thera renin nt ae
invested or Dat dir gen
‘hedirectory can he viewed as a symbol table ha tranalaes eames
ther directory entries
take such a view, we see thatthe irecoryitslfcan be organi’
| famany ways
organization must be allowed to insert entries, ta delet nti 12
ach fora namned entry, and to ist ll he entries i the dineetors.
ull
Scanned by CamScanner
Seamed wth camscamer Scanned by CamScanner
Seamed wth camscamer advent of the Internet, a file transfer y
Protocol (FTP) has become widely-used.
abe used to access (read and
IP sites offer public file sharing orat
by downloading them, using a pu
be “anonymous”),
Scanned by CamScanner
Seamed wth camscamer ier of the file up to maxintum Hit mast be knwwn atthe
ime sy
idsk space is wasted due to fragmentation. ‘
sito remove fragmentation. a
allocation : al
inlined allocation, each file ie alinked it oink Mocs: the
Wocks may be seattered anywhere onthe disk. 7
Medirectory contains a pointer tothe rst and last locks the
=
example, a file o ive blocks might tart at ble 9and continue
tblock 16, then block Meee ec 0, and finaly Wok *
block contains a pointer to the next block.
‘pointers are not made available tothe user.
hus, i each block is 512 bytes in size, and a
) requires 4 bytes, then the user see
Scanned by CamScanner
Seamed wth camscamer entry for every recordin the main fle
‘Rta inteaforwutned as a seqential Se for eumeteehing
‘pertains contain enti to record where the felieoae
ent
ans 77s of variable length, some record wil no conain al
fields
Thema new record is added othe main fl, ll he index flemst
be updated,
Fig. 6.22.2 shows the index with relative Mes
Indexed fle are used mestly in applications where timelines of
information are
