Operating Systems 23CDI304

MODULE: 4
CONTENTS:
 MemoryManagement:

 Memorymanagementstrategies:Background;
 Swapping;
 Contiguousmemoryallocation;
 Paging;
 Structureofpagetable;
 Segmentation.

 VirtualMemoryManagement:

 Background;
 Demandpaging;
 Copy-on-write;
 Pagereplacement;
 Allocationofframes;
 Thrashing

1 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

MEMORYMANAGEMENT
MainMemoryManagementStrategies

 Everyprogramtobeexecutedhastobeexecutedmustbeinmemory.Theinstructionmustbefetch
edfrommemorybeforeitisexecuted.
 Inmulti-taskingOSmemorymanagementiscomplex, because asprocesses
areswappedinandout of theCPU, theircodeanddata mustbeswappedinandoutof
memory.

BasicHardware

 Main memory, cache and CPU registers in the processors are the only
storagespacesthatCPUcanaccessdirectly.
 The program and data must be bought into the memory from the disk, for the process
torun. Each process has a separate memory space and must access only this range of
legaladdresses. Protection of memory is required to ensure correct operation. This
preventionisprovidedbyhardwareimplementation.
 Two registers are used - a base register and a limit register. The base register holds
thesmallestlegalphysicalmemoryaddress;thelimitregisterspecifiesthesizeoftherange.
 For example, The base register holds the smallest legal physical memory address;
thelimit register specifies the size of the range. For example, if the base register
holds300040 and limit register is 120900, then the program can legally access all
addressesfrom300040through420940(inclusive).

Figure:A baseandalimit-register definealogical-addressspace

 The base and limit registers can be loaded only by the operating system, which uses
aspecial privilegedinstruction. Since privileged instructions can beexecutedonly
inkernelmodeonlytheoperatingsystemcanloadthebaseandlimitregisters.

2 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Figure:Hardwareaddressprotectionwith baseandlimit-registers

AddressBinding
 User programs typically refer to memory addresses with symbolic names. These
symbolicnamesmustbemappedorboundtophysicalmemoryaddresses.
 Addressbindingof instructionsto memory-addressescanhappenat3 differentstages.

1. Compile Time- If it is known at compile time where a program will reside in

physicalmemory,thenabsolutecodecanbegeneratedbythecompiler,containingactualphysic
al addresses. However, if the load address changes at some later time, then
theprogramwillhavetoberecompiled.
2. Load Time- If the location at which a program will be loaded is not known at
compiletime,thenthecompilermust
generaterelocatablecode,whichreferencesaddressesrelative to the start of the program. If
that starting address changes, then the programmustbereloadedbutnotrecompiled.
3. Execution Time- If a program can be moved around in memory during the course of
itsexecution,then bindingmustbe delayed untilexecutiontime.

Figure:Multistep processingof auserprogram

3 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

LogicalVersusPhysicalAddressSpace

 TheaddressgeneratedbytheCPUisalogicaladdress,whereasthememoryaddresswhereprogra
msareactuallystoredisaphysicaladdress.
 Thesetofalllogicaladdressesusedbyaprogramcomposesthelogicaladdressspace,andthesetof
allcorrespondingphysicaladdressescomposesthe physicaladdress space.
 Theruntimemappingoflogicaltophysicaladdressesishandledbythememory-
managementunit (MMU).
 Oneofthesimplestisa modificationofthebase-registerscheme.
 Thebaseregisteristermedarelocationregister
 Thevalueintherelocation-registerisaddedtoeveryaddressgeneratedbyauser-
processatthetimeitissent tomemory.
 Theuser-programdealswithlogical-addresses;itneverseestherealphysical-
addresses.

Figure:Dynamicrelocationusing a relocation-register

DynamicLoading
 Thiscanbe usedto obtainbetter memory-spaceutilization.
 Aroutineisnotloadeduntilitiscalled.

Thisworksasfollows:
1. Initially,all routinesarekeptondiskina relocatable-loadformat.
2. Firstly,themain-programisloadedinto memoryandisexecuted.
3. When a main-program calls the routine, the main-program first checks to see
whether theroutinehasbeenloaded.
4. Ifroutinehasbeennotyetloaded,theloaderiscalledtoloaddesiredroutineintomemory.
5. Finally,controlispassedtothenewlyloaded-routine.

Advantages:
1. Anunused routineisneverloaded.
2. Usefulwhenlargeamountsof codeareneededtohandleinfrequentlyoccurringcases.
3. Although the total program-size may be large, the portion that is used (and hence
loaded)maybemuchsmaller.
4. DoesnotrequirespecialsupportfromtheOS.

4 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

DynamicLinkingandSharedLibraries

 With static linkinglibrary modules get fully included in executable modules,

wastingboth disk space and main memory usage, because every program that included a
certainroutine from the library would have to have their own copy of that routine linked
intotheirexecutablecode.
 With dynamiclinking, however, only a stub is linked into
theexecutablemodule,containingreferences to theactuallibrarymodulelinkedin atrun time.
 The stub is a small piece of code used to locate the appropriate memory-
residentlibrary-routine.
 This method saves disk space, because the library routines do not need to be
fullyincludedintheexecutablemodules,onlythestubs.
 An added benefit of dynamically linked libraries (DLLs, also known
as sharedlibrariesor
sharedobjectsonUNIXsystems)involveseasyupgrades andupdates.

Sharedlibraries
 Alibrarymaybereplacedbyanewversion,andallprogramsthatreferencethelibrarywillautoma
ticallyusethenewone.
 Versioninfo.isincludedinbothprogram&librarysothatprogramswon'taccidentallyexecutein
compatibleversions.

Swapping
 Aprocess mustbe loadedintomemoryin ordertoexecute.
 If there is not enough memory available to keep all running processes in memory at
thesame time, then some processes that are not currently using the CPU may have
theirmemoryswappedouttoafastlocaldiskcalledthebackingstore.
 Swapping is the process of moving a process from memory to backing store and
movinganotherprocessfrombackingstoretomemory.Swappingisaveryslowprocesscompare
dtootheroperations.
 A variant ofswapping policy isused forpriority-based scheduling algorithms.Ifahigher-
priority process arrives and wants service, the memory manager can swap out thelower-
priority process and then load and execute the higher-priority process. When thehigher-
priority process finishes, the lower-priority process can be swapped back in
andcontinued. This variantofswappingiscalledrollout,rollin.

Swappingdepends uponaddress-binding:
 Ifbindingisdoneatload-time,thenprocesscannotbeeasilymovedtoadifferentlocation.
 Ifbindingisdoneatexecution-time,thenaprocesscanbeswappedintoadifferentmemory-
space,becausethephysical-addressesarecomputedduring execution-time.

Majorpartofswap-timeistransfer-time;i.e.totaltransfer-
timeisdirectlyproportionaltotheamountofmemoryswapped.

5 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Disadvantages:
1. Context-switchtimeisfairlyhigh.
2. If we want to swap a process, we must be sure that it is
completely idle.Twosolutions:
i) NeverswapaprocesswithpendingI/O.
ii) ExecuteI/OoperationsonlyintoOSbuffers.

Figure:Swappingoftwoprocessesusingadiskasabackingstore

Example:
Assume that the user processis 10MB in sizeandthe backingstore is astandard harddisk
withatransfer rate of40 MB per second.
The actual transfer of the 10-MB process to or from main memory
takes10000KB/40000KBpersecond=1/4second
=250 milliseconds.
Assuming that no head seeks are necessary, and assuming an average latency of 8
milliseconds,the swap time is 258 milliseconds. Since we must both swap out and swap in, the
total swaptimeisabout516milliseconds.

6 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

ContiguousMemoryAllocation

 The main memory must accommodate both the operating system and the various
userprocesses. Therefore we need to allocate the parts of the main memory in the
mostefficientwaypossible.
 Memory is usually divided into 2 partitions: One for the resident OS. One for the
userprocesses.
 Eachprocessiscontainedinasinglecontiguoussection of memory.

1. MemoryMappingandProtection

 Memory-protectionmeansprotectingOSfromuser-processandprotectinguser-
processesfromoneanother.
 Memory-protectionisdoneusing
o Relocation-register:containsthevalueofthe smallestphysical-address.
o Limit-register:contains therangeoflogical-addresses.
 Eachlogical-addressmustbeless thanthelimit-register.
 TheMMUmapsthelogical-addressdynamicallybyaddingthevalueintherelocation-
register.Thismapped-address issent tomemory
 WhentheCPUschedulerselectsaprocessforexecution,thedispatcherloadstherelocationandli
mit-registerswiththecorrectvalues.
 BecauseeveryaddressgeneratedbytheCPUischeckedagainsttheseregisters,wecanprotectthe
OSfromtherunning-process.
 Therelocation-
registerschemeprovidesaneffectivewaytoallowtheOSsizetochangedynamically.
 TransientOScode:Codethatcomes&goesasneededtosavememory-
spaceandoverheadforunnecessaryswapping.

Figure:Hardwaresupportforrelocationandlimit-registers

7 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

2. MemoryAllocation

Twotypesof memorypartitioningare:

1. Fixed-sizedpartitioning
2. Variable-sized

partitioning1.Fixed-

sizedPartitioning

 Thememoryisdividedintofixed-sizedpartitions.
 Eachpartitionmaycontainexactlyoneprocess.
 Thedegreeofmultiprogramming isboundbythenumberofpartitions.
 Whenapartitionisfree,aprocessisselectedfromtheinputqueueandloadedintothefreepartition.
 Whentheprocessterminates,thepartition becomesavailableforanotherprocess.

2. Variable-sizedPartitioning

 TheOSkeepsatableindicating
whichpartsofmemoryareavailable andwhichpartsareoccupied.
 Aholeisablockofavailablememory.Normally,memorycontainsasetofholesofvarioussizes.
 Initially,allmemoryis availableforuser-processesandconsideredonelargehole.
 Whenaprocess arrives,theprocessisallocated memoryfromalargehole.
 Ifwefindthehole,weallocateonlyasmuchmemoryasisneededandkeeptheremaining
memoryavailabletosatisfyfuturerequests.

Threestrategiesusedtoselectafreeholefromthesetofavailableholes:

1. FirstFit:Allocatethefirstholethatisbigenough.Searchingcanstarteitheratthebeginning
ofthesetofholesoratthelocationwherethepreviousfirst-fitsearchended.

2. BestFit:Allocatethesmallestholethatisbigenough.Wemustsearchtheentirelist,unlessthelisti
sorderedbysize.Thisstrategyproducesthesmallestleftoverhole.

3. WorstFit:Allocatethelargesthole.Again,wemustsearchtheentirelist,unlessitissortedbysize.
Thisstrategyproducesthelargestleftoverhole.

First-fitandbestfitarebetterthanworstfitin termsof decreasingtimeandstorageutilization.

8 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

3. Fragmentation

Twotypesof memoryfragmentation:
1. Internalfragmentation
2. Externalfragmentation

1. InternalFragmentation
 Thegeneralapproachistobreakthephysical-memoryintofixed-sizedblocksandallocate
memoryinunits basedonblock size.
 Theallocated-memorytoaprocessmaybeslightlylargerthantherequested-memory.
 Thedifferencebetweenrequested-memoryandallocated-
memoryiscalledinternalfragmentationi.e.Unusedmemorythatisinternaltoapartition.

2. External Fragmentation
 External fragmentation occurs when there is enough total memory-space to satisfy
arequestbuttheavailable- spacesarenotcontiguous.
(i.e.storageisfragmentedintoalargenumberofsmallholes).
 Boththefirst-fitandbest-fitstrategiesformemory-
allocationsufferfromexternalfragmentation.
 Statistical analysis of first-fit reveals that given N allocated blocks, another 0.5 N
blockswillbelosttofragmentation.This propertyisknownasthe50-percentrule.

Twosolutionstoexternalfragmentation:
 Compaction:Thegoalistoshufflethememory-contentstoplaceallfreememorytogether in one
large hole. Compaction is possible only if relocation is dynamic and doneatexecution-
time
 Permit the logical-address space of the processes to be non-contiguous. This allows
aprocesstobeallocatedphysical-
memorywhereversuchmemoryisavailable.Twotechniquesachievethissolution:
1)Pagingand2)Segmentation.

9 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Paging

 Pagingisamemory-managementscheme.
 Thispermitsthephysical-addressspaceofaprocesstobe non-contiguous.
 Thisalsosolvestheconsiderableproblemoffittingmemory-
chunksofvaryingsizesontothebacking-store.
 Traditionally: Support forpaginghasbeen handledbyhardware.
 Recentdesigns:Thehardware&OSarecloselyintegrated.

BasicMethodofPaging

 The basic method for implementing paging involves breaking physical memory
intofixed-sized blocks called frames and breaking logical memory into blocks of the
samesizecalled pages.
 When a process is to be executed, its pages are loaded into any available memory
framesfromthebackingstore.
 The backing store is divided into fixed-sized blocks that are of the same size as
thememoryframes.

Thehardwaresupportforpagingisillustrated inFigure1.

Figure1:Paging hardware

 AddressgeneratedbyCPUisdividedinto2parts(Figure 2):
1. Page-number (p) is used as an index to the page-table. The page-table
contains thebase-addressofeachpageinphysical-memory.
2. Offset(d)iscombinedwiththebase-addresstodefinethephysical-
address.Thisphysical-addressissenttothememory-unit.
 Thepagetable mapsthepagenumbertoaframenumber,toyielda physicaladdress
 Thepagetablemapsthepagenumbertoaframenumber,toyieldaphysicaladdresswhichalsohast
woparts:Theframenumberandtheoffsetwithinthatframe.
 Thenumberofbitsintheframenumberdetermineshowmanyframesthesystemcanaddress,and
thenumber ofbitsin theoffsetdetermines the sizeofeachframe.

10 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Thepaging model ofmemoryisshownin Figure2.

Figure2: Pagingmodeloflogicalandphysical memory.

 Thepagesize(liketheframesize)isdefinedbythehardware.
 The size of a page is typically a power of 2, varying between 512 bytes and 16 MB
perpage,dependingonthecomputerarchitecture.
 The selection of a power of 2 as a page size makes the translation of a logical
addressintoapagenumberandpageoffset.
 If the size of logical address space is 2m and a page size is 2n addressing units (bytes
orwords), then the high-order m – n bits of a logical address designate the page
number,andthe nlow-orderbitsdesignatethepageoffset.

Thus,thelogicaladdressisasfollows:
pagenumber pageoffset
p d
m-n n

 When a process requests memory (e.g. when its code is loaded in from disk), free
framesareallocated fromafree-framelist, andinsertedintothatprocess'spagetable.
 Processes are blocked from accessing anyone else's memory because all of their
memoryrequests are mapped through their page table. There is no way for them to
generate anaddressthatmaps into anyother process'smemoryspace.
 The operating system must keep track of each individual process's page table, updating
itwhenever the process's pages get moved in and out of memory, and applying the
correctpage table when processing system calls for a particular process. This all
increases theoverheadinvolvedwhenswappingprocessesinandoutoftheCPU.

11 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Figure:Freeframes(a)beforeallocationand(b)afterallocation.

HardwareSupport

TranslationLookasideBuffer

 Aspecial,small,fastlookuphardwarecache,calledatranslationlook-asidebuffer(TLB).
 EachentryintheTLBconsistsoftwoparts:akey(ortag)andavalue.
 When the associative memory is presented with an item, the item is compared with
allkeys simultaneously. If the item is found, the corresponding value field is returned.
Thesearch is fast; the hardware, however, is expensive. Typically, the number of entries
in aTLBissmall,oftennumberingbetween64and1,024.
 TheTLBcontains onlyafewofthepage-tableentries.

Working:
 Whenalogical-addressisgeneratedbytheCPU,itspage-numberispresentedtotheTLB.
 If the page-number is found (TLB hit), its frame-number is immediately available
andusedtoaccessmemory
 If page-number is not in TLB (TLB miss), a memory-reference to page table must
bemade.Theobtainedframe-numbercanbeusedtoaccessmemory(Figure1)

12 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Figure1:Paging hardwarewithTLB

 Inaddition,weaddthepage-numberandframe-
numbertotheTLB,sothattheywillbefoundquicklyonthenextreference.
 If theTLBisalreadyfullofentries,theOSmustselectoneforreplacement.
 Percentageoftimesthataparticularpage-numberisfoundinthe TLB iscalledhitratio.

Advantage: Search operation is

fast.Disadvantage:Hardwareisexpensive.

 SomeTLBshavewireddownentriesthatcan'tberemoved.
 SomeTLBsstoreASID(address-
spaceidentifier)ineachentryoftheTLBthatuniquelyidentifyeachprocessandprovideaddresss
paceprotectionforthatprocess.

Protection

 Memory-protection isachievedbyprotection-bitsforeachframe.
 Theprotection-bitsarekeptinthe page-table.
 Oneprotection-bitcandefineapagetoberead-writeorread-only.
 Everyreferencetomemorygoesthroughthepage-tabletofindthecorrectframe-number.
 Firstly,thephysical-addressiscomputed.Atthesametime,theprotection-
bitischeckedtoverifythatnowritesarebeingmadetoaread-onlypage.
 Anattempttowritetoaread-onlypagecausesahardware-
traptotheOS(ormemoryprotectionviolation).

13 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

ValidInvalid Bit
 Thisbitisattachedtoeachentryin thepage-table.
 Validbit:“valid”indicatesthattheassociatedpageisintheprocess’logicaladdressspace,andist
husalegalpage
 Invalidbit: “invalid”indicatesthatthepageis notinthe process’logicaladdressspace

Illegaladdressesaretrappedbyuseofvalid-invalidbit.
TheOSsetsthisbitforeachpagetoallowor disallowaccesstothepage.

Figure:Valid(v)orinvalid(i)bitinapage-table

SharedPages

 Anadvantageof pagingisthepossibilityof sharingcommoncode.

 Re-entrantcode(PureCode)isnon-self-modifyingcode,itneverchangesduringexecution.
 Twoormoreprocesses canexecutethesamecodeatthesametime.
 Eachprocesshasitsowncopyofregistersanddata-
storagetoholdthedatafortheprocess'sexecution.
 Thedatafor2differentprocesseswillbedifferent.
 Onlyone copyoftheeditor need bekept inphysical-memory(Figure 5.12).
 Each user's page-table maps onto the same physical copy of the editor, but data pages
aremappedontodifferentframes.

Disadvantage:
Systemsthatuseinvertedpage-tableshavedifficultyimplementingshared-memory.

14 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Figure:Sharingofcodeinapagingenvironment

Structureof thePage Table

Themostcommon techniquesforstructuringthepagetable:
1. HierarchicalPaging
2. HashedPage-tables
3. InvertedPage-tables

1. HierarchicalPaging
 Problem:Mostcomputerssupportalargelogical-
addressspace(232to264).Inthesesystems,thepage-tableitselfbecomesexcessivelylarge.
 Solution:Dividethepage-tableintosmallerpieces.

TwoLevelPagingAlgorithm:
 Thepage-tableitselfis alsopaged.
 Thisisalsoknownasaforward-mappedpage-
tablebecauseaddresstranslationworksfromtheouterpage-tableinwards.

15 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Figure:Atwo-levelpage-tablescheme
Forexample:
Consider the system with a 32-bit logical-address space and a page-size of 4
KB.Alogical-addressisdividedinto
→20-bitpage-numberand
→12-bitpage-offset.
Sincethepage-tableispaged,thepage-numberisfurtherdividedinto
→10-bitpage-numberand
→10-bitpage-offset.

Thus,alogical-address isasfollows:

 wherep1isanindexintotheouterpagetable,andp2isthedisplacementwithinthepageofthei
nnerpagetable

The address-translation method for this architecture is shown in below figure. Because
addresstranslation works from the outer page table inward, this scheme is also known as a
forward-mappedpagetable.

Figure:Addresstranslationforatwo-level32-bitpagingarchitecture

16 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

2. HashedPage Tables
 Thisapproachisusedforhandlingaddressspaceslargerthan32bits.
 Thehash-valueisthevirtualpage-number.
 Eachentryinthehash-tablecontainsalinked-
listofelementsthathashtothesamelocation(tohandlecollisions).
 Eachelementconsists of 3fields:
1. Virtualpage-number
2. Valueof the mappedpage-frameand
3. Pointerto thenextelementinthe linked-list.

Thealgorithmworksasfollows:
 Thevirtual page-numberishashed intothehash-table.
 Thevirtual page-numberiscomparedwith thefirstelementinthelinked-list.
 Ifthereisamatch,thecorrespondingpage-frame(field2)isusedtoformthedesiredphysical-
address.
 Ifthereisnomatch,subsequententriesinthelinked-listaresearchedforamatchingvirtualpage-
number.

Figure:Hashedpage-table

3. InvertedPage Tables
 Hasoneentryforeach realpageofmemory.
 Eachentryconsistsofvirtual-addressofthepagestoredinthatrealmemory-
locationandinformationabouttheprocessthatownsthepage.
 Eachvirtual-addressconsistsof atriplet<process-id,page-number,offset>.
 Eachinvertedpage-tableentryisapair<process-id,page-number>

17 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Figure:Invertedpage-table

Thealgorithmworksasfollows:
1. Whenamemory-referenceoccurs,partofthevirtual-address,consistingof<process-id,page-
number>,ispresentedtothememorysubsystem.
2. Theinverted page-tableisthensearchedforamatch.
3. Ifamatchisfound,at entryi-then thephysical-address<i, offset> isgenerated.
4. If nomatchisfound,thenanillegal addressaccesshasbeenattempted.

Advantage:
1. Decreasesmemoryneededtostore eachpage-table

Disadvantages:
1. Increasesamountof time neededtosearchtablewhenapagereferenceoccurs.
2. Difficultyimplementing shared-memory

Segmentation

BasicMethodofSegmentation
 Thisisa memory-managementschemethatsupportsuser-viewofmemory(Figure1).
 Alogical-addressspaceisacollectionof segments.
 Eachsegmenthasa name andalength.
 Theaddressesspecifybothsegment-nameandoffsetwithinthesegment.
 Normally,theuser-
programiscompiled,andthecompilerautomaticallyconstructssegmentsreflectingtheinputpr
ogram.
 Forex:Thecode,Globalvariables,Theheap,fromwhichmemoryisallocated,Thestacksusedby
eachthread,ThestandardClibrary

18 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Figure:Programmer’s viewofaprogram

HardwaresupportforSegmentation
 Segment-tablemaps 2dimensionaluser-defined addresses intoone-
dimensionalphysicaladdresses.
 Inthesegment-table,eachentryhasfollowing2fields:
1. Segment-basecontainsstartingphysical-addresswherethesegmentresidesinmemory.
2. Segment-limitspecifiesthelengthofthesegment(Figure2).
 Alogical-addressconsistsof 2parts:
1. Segment-number(s) is usedasanindexto the segment-table
2. Offset(d)mustbebetween0andthesegment-limit.
 Ifoffsetisnotbetween0&segment-limit,thenwetraptotheOS(logical-
addressingattemptbeyondendofsegment).
 Ifoffsetislegal,thenitisaddedtothesegment-basetoproducethephysical-memoryaddress.

Figure:Segmentation hardware

VIRTUALMEMORYMANAGEMENT
19 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Background

 Virtual memoryis atechniquethatallowsfortheexecution of partiallyloadedprocess.

 Advantages:
 A program will not be limited by the amount of physical memory that is
availableusercanabletowriteintolargevirtualspace.
 Sinceeachprogramtakeslessamountofphysicalmemory,morethanoneprogram
couldbe runat thesametimewhich canincreasethe
throughputandCPUutilization.
 Less i/o operation is needed to swap or load user program in to memory. So
eachuserprogramcouldrunfaster.

Fig:Virtualmemorythatislargerthanphysicalmemory.

 Virtual memory is the separation of users logical memory from physical memory.
Thisseparation allows an extremely large virtual memory to be provided when these is
lessphysicalmemory.
 Separating logical memory from physical memory also allows files and memory to
besharedbyseveraldifferentprocessesthroughpagesharing.

20 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Fig:SharedLibraryusingVirtual Memory

 Virtualmemoryis implementedusingDemandPaging.
 Virtual address space: Every process has a virtual address space i.e used as the stack
orheapgrowsin size.

Fig:Virtualaddressspace
DEMANDPAGING
 A demand paging is similar to paging system with swapping when we want to execute
aprocess we swap the process the in to memory otherwise it will not be loaded in
tomemory.
 A swapper manipulates the entire processes, where as a pager manipulates
individualpagesofthe process.
 Bringapageintomemoryonlywhenit isneeded
 LessI/Oneeded
 Lessmemoryneeded
 Fasterresponse

21 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

 Moreusers
 Pageisneeded⇒referenceto it
 invalidreference⇒abort
 not-in-memory⇒bringtomemory
 Lazyswapper–neverswapsapageintomemoryunlesspagewillbeneeded
 Swapperthatdealswithpagesisapager.

Fig:Transferofapaged memoryintocontinuousdiskspace

 Basic concept: Instead of swapping the whole process the pager swaps only the
necessarypages in to memory. Thus it avoids reading unused pages and decreases the swap
time andamountofphysicalmemoryneeded.
 The valid-invalid bit scheme can be used to distinguish between the pages that are on the
diskandthatareinmemory.
 Witheachpagetableentryavalid–invalidbitisassociated
 (v⇒in-memory,i⇒not-in-memory)
 Initiallyvalid–invalidbitissettoionallentries
 Exampleof apagetablesnapshot:

 Duringaddresstranslation,ifvalid–invalidbitinpagetableentryisI⇒pagefault.
 Ifthebitisvalidthenthepageisbothlegalandisinmemory.
 Ifthebitisinvalidtheneitherpageisnotvalid orisvalidbutiscurrentlyonthedisk. Marking

22 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

a page as invalid will have no effect if the processes never access to that page. Suppose if
itaccess the page which is marked invalid, causes a page fault trap. This may result in failure
ofOStobringthedesiredpageintomemory.

Fig:PageTable whensome pages arenot in mainmemory

PageFault

Ifapageisneededthatwas notoriginallyloaded up,then apagefaulttrapisgenerated.

Stepsin HandlingaPageFault
1. Thememory addressrequestedis first checked,tomakesureitwasavalidmemoryrequest.
2. Ifthereferenceistoaninvalidpage,theprocessisterminated.Otherwise,ifthepageisnotpresentin
memory,itmustbepagedin.
3. Afreeframeislocated, possiblyfromafree-framelist.
4. Adiskoperationisscheduledtobringinthenecessarypagefromdisk.
5. After the page is loaded tomemory, the process's page table is updated with thenewframe
number, and the invalid bit is changed to indicate that this is now a valid pagereference.
6. Theinstructionthatcausedthepagefaultmustnowberestartedfromthebeginning

23 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Fig:stepsinhandlingpagefault

PureDemandPaging:Neverbringapageintomainmemoryuntilit isrequired.
 We canstartexecuting aprocesswithoutloadinganyofitspagesinto mainmemory.
 Pagefaultoccursforthenonmemoryresidentpages.
 Afterthepageisbroughtintomemory,processcontinues toexecute.
 Againpagefaultoccursforthenextpage.

Hardwaresupport:Fordemandpagingthesamehardwareisrequiredaspagingandswapping.
1. Pagetable:-Has theabilitytomarkanentryinvalidthroughvalid-invalid bit.
2. Secondarymemory:-Thisholdsthepagesthat arenotpresent in main memory.

PerformanceofDemandPaging:Demandpagingcanhavesignificanteffectontheperformanceoftheco
mputer system.
 LetPbetheprobabilityofthepagefault(0<=P<=1)
 Effectiveaccesstime=(1-P)*ma+P *pagefault.
 WhereP=pagefaultand ma=memoryaccesstime.
 Effectiveaccesstimeisdirectlyproportionaltopagefaultrate.Itisimportanttokeeppagefaultratel
owindemandpaging.
DemandPagingExample
 Memoryaccesstime=200 nanoseconds
 Averagepage-faultservicetime=8milliseconds
 EAT=(1–p)x200+p(8milliseconds)
=(1 –px200+px8,000,000
=200+ px 7,999,800
 Ifoneaccessoutof1,000causesapagefault,thenEAT=8.2microseconds.Thisisaslowdownby
afactorof40.

24 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

COPY-ON-WRITE
 Technique initially allows the parent and the child to share the
samepages.Thesepagesaremarkedascopyon-
writepagesi.e.,ifeitherprocesswritestoasharedpage,acopyofsharedpageiscreated.
 Eg:-If a child process try to modify a page containing portions of the stack; the
OSrecognizes them as a copy-on-write page and create a copy of this page and maps it
onto the address space of the child process. So the child process will modify its
copiedpageandnotthepagebelongingtoparent.Thenewpagesareobtainedfromthepooloffre
epages.
 The previous contents of pages are erased before getting them into main memory.
ThisiscalledZero –onfill demand.

a) BeforeProcess1modifiespageC

b) Afterprocess1modifiespageC

25 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

PAGEREPLACEMENT
 Page replacement policy deals with thesolution ofpages inmemory to be replacedby
anewpagethat must bebroughtin.Whenauserprocessisexecutingapagefaultoccurs.
 Thehardwaretrapstotheoperatingsystem,whichcheckstheinternaltabletoseethatthisisapage
faultandnotanillegalmemoryaccess.
 The operating system determines where the derived page is residing on the disk, and
thisfindsthattherearenofreeframesonthelistoffreeframes.
 When all the frames are in main memory, it is necessary to bring a new page to satisfy
thepage fault, replacement policy is concerned with selecting a page currently in memory to
bereplaced.
 Thepagei,etoberemovedshouldbethe pagei,eleastlikelytobe referencedinfuture.

Fig:PageReplacement

WorkingofPage ReplacementAlgorithm
1. Findthelocationof derivedpageonthe disk.
2. FindafreeframexIfthereisafreeframe,useit.xOtherwise,useareplacementalgorithmto
selectavictim.
 Writethevictimpagetothedisk.
 Changethepageandframetablesaccordingly.
3. Readthedesiredpageintothefreeframe;changethepageandframetables.
4. Restarttheuserprocess.

VictimPage
 Thepagethatissupportedoutof physicalmemoryiscalledvictimpage.
 Ifno
framesarefree,thetwopagetransformscome(outandonein)areread.Thiswillseetheeffectiveacc
esstime.
 Eachpageorframemayhaveadirty(modify)bitassociatedwiththehardware.Themodifybitforap
ageissetbythehardwarewheneveranywordorbyteinthepageis

26 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

writteninto,indicating thatthepagehasbeenmodified.
 When we select the page for replacement, we check its modify bit. If the bit is set, then
thepageismodifiedsinceitwasreadfromthedisk.
 If the bit was not set, the page has not been modified since it was read into
memory.Therefore, if the copy of the page has not been modified we can avoid writing the
memorypagetothedisk,if itisalreadythere.Sumpagescannotbemodified.

Modifybit/Dirty bit:
 Eachpage/framehas a modifybitassociated withit.
 Ifthepageisnotmodified(read-only)thenonecandiscardsuchpagewithoutwritingitontothedisk.
Modifybitofsuchpageissetto0.
 Modifybitissetto1,ifthepagehasbeenmodified.Suchpagesmustbewrittentothedisk.
 Modifybitisusedtoreduceoverheadofpagetransfers–onlymodifiedpagesarewrittentodisk

PAGEREPLACEMENTALGORITHMS
 Wantlowestpage-faultrate
 Evaluatealgorithmbyrunningitonaparticularstringofmemoryreferences(referencestring)andco
mputingthenumberofpagefaultsonthatstring
 Inallourexamples,thereferencestringis
1, 2, 3, 4, 1, 2, 5, 1, 2, 3, 4, 5

FIFOAlgorithm:
 Thisisthesimplestpagereplacementalgorithm.AFIFOreplacementalgorithmassociateseachpa
ge thetimewhen thatpagewasbrought intomemory.
 WhenaPageisto bereplacedtheoldestoneis selected.
 Wereplacethequeueattheheadofthequeue.Whenapageisbroughtintomemory,weinsertitattheta
ilofthequeue.
 Inthefollowingexample,areferencestringisgivenandthereare3freeframes.Thereare20pagereq
uests,whichresultsin15page faults

27 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Belady’sAnomaly

 Forsomepagereplacementalgorithm,thepagefaultmayincreaseasthenumberofallocatedframes

moreframes⇒ morepagefaults
increases.FIFOreplacement algorithmmayfacethisproblem.

Example: Consider thefollowingreferencesstringwithframesinitiallyempty.

 Thefirstthreereferences(7,0,1)casespagefaultsandarebroughtintotheemptyframes.
 Thenextreferences2replacespage7becausethepage7wasbroughtinfirst.xSince0isthenext
referencesand0isalreadyinmemoryehasnopagefaults.
 Thenextreferences3resultsinpage0beingreplacedsothatthenextreferencesto0causerpagefault.
This willcontinuetilltheendof string.Thereare 15faultsalltogether.

FIFOIllustratingBelady’sAnomaly

28 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

OptimalAlgorithm
 Optimalpagereplacementalgorithmis mainlytosolvetheproblemofBelady’sAnomaly.
 Optimalpagereplacementalgorithmhasthelowestpagefaultrateofallalgorithms.
 Anoptimalpage replacementalgorithmexists andhasbeencalledOPT.

The working is simple “Replace the page that will not be used for the longest period of
time”Example:considerthefollowingreferencestring
 Thefirstthreereferencescausefaultsthatfillthethreeemptyframes.
 The references to page 2 replaces page 7, because 7 will not be used until reference 18.
xThepage0willbeusedat5andpage1at 14.
 With only 9 page faults, optimal replacement is much better than a FIFO, which had
15faults. This algorithm is difficult t implement because it requires future knowledge
ofreferencestrings.
 Replacepagethatwill notbeusedforlongest periodoftime

OptimalPageReplacement

LeastRecently Used(LRU)Algorithm
 TheLRU(LeastRecentlyUsed)algorithm,predictsthatthepagethathasnotbeenusedinthelonge
sttimeistheonethatwillnot beusedagaininthenearfuture.
 SomeviewLRUasanalogoustoOPT,butherewelookbackwardsintimeinsteadofforwards.

ThemainproblemtohowtoimplementLRU istheLRUrequiresadditionalh/wassistance.

29 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

Twoimplementationarepossible:
1. Counters: In this we associate each page table entry a time -of -use field, and add
tothecpualogicalclockorcounter.Theclockisincrementedforeachmemoryreference.
When a reference to a page is made, the contents of the clock register
arecopiedtothetime-of-usefieldinthepagetableentryforthatpage.Inthiswaywehave the
time of last reference to each page we replace the page with smallest
timevalue.Thetimemustalsobemaintainedwhenpagetablesarechanged.
2. Stack: Another approach to implement LRU replacement is to keep a stack of
pagenumbers when a page is referenced it is removed from the stack and put on to the
topof stack. In this way the top of stack is always the most recently used page and
thebottom in least recently used page. Since the entries are removed from the stack it
isbestimplementbyadoublylinkedlist.Withaheadandtailpointer.

Note: Neither optimal replacement nor LRU replacement suffers from Belady’s
Anamoly.Thesearecalledstackalgorithms.

LRU-ApproximationPageReplacement
 Manysystemsoffersomedegreeofhardwaresupport,enoughto approximateLRU.
 Inparticular,manysystemsprovideareferencebitforeveryentryinapagetable,whichis set
anytime that page is accessed. Initially all bits are set to zero, and they can also all
becleared at any time. One bit distinguishes pages that have been accessed since the last
clearfromthosethathavenotbeenaccessed.
Additional-Reference-BitsAlgorithm
 An8-bitbyte (referencebit)isstoredforeachpageinatableinmemory.
 At regular intervals (say, every 100 milliseconds), a timer interrupt transfers control to
theoperating system. The operating system shifts the reference bit for each page into
thehigh-order bit of its 8-bit byte, shifting the other bits right by 1 bit and discarding the
low-orderbit.
 These8-bitshiftregisterscontainthehistoryof pageuseforthelasteighttimeperiods.
 If the shift register contains 00000000, then the page has not been used for eight
timeperiods.
 A page with a history register value of 11000100 has been used more recently than
onewithavalueof01110111.
Second-chance(clock)pagereplacement algorithm
 ThesecondchancealgorithmisaFIFOreplacementalgorithm,exceptthereferencebitisusedto
givepagesasecondchanceatstayinginthepagetable.
 Whenapagemustbereplaced,thepagetableisscannedinaFIFO(circularqueue)manner.
 Ifapageisfoundwithitsreferencebitas‘0’,thenthatpageisselectedasthenextvictim.

30 Dept.of CSE(DS),SJBIT
 Operating Systems 23CDI304

 If the reference bitvalueis‘1’, then the page is given a second chance and its reference
bitvalueiscleared (assigned as‘0’).
 Thus, a page that is given a second chance will not be replaced until all other pages
havebeen replaced (or given second chances). In addition, if a page is used often, then it
setsitsreferencebitagain.
 Thisalgorithmisalsoknownastheclockalgorithm.

EnhancedSecond-ChanceAlgorithm
 Theenhancedsecond chancealgorithmlooksatthereferencebitandthe
modifybit(dirtybit)asanorderedpage,andclassifiespagesintooneoffourclasses:
1. (0,0)-Neitherrecentlyusednormodified.
2. (0,1)-Notrecentlyused,butmodified.
3. (1,0)-Recentlyused,butclean.
4. (1,1)-Recentlyusedandmodified.
 This algorithm searches the page table in a circular fashion, looking for the first page it
canfind in the lowest numbered category. i.e. it first makes a pass looking for a ( 0, 0 ), and
thenifitcan'tfindone, itmakes anotherpasslookingfor a(0,1),etc.
 The main difference between this algorithm and the previous one is the preference
forreplacingcleanpagesifpossible.

CountBasedPageReplacement
Thereismanyotheralgorithmsthatcanbeusedforpagereplacement,wecankeepacounterofthenumb
erofreferencesthathasmadetoapage.
a) LFU(leastfrequentlyused):
This causes the page with the smallest count to be replaced. The reason for this
selectionisthatactivelyusedpageshouldhavealargereferencecount.

31 Dept.of CSE(DS),SJBIT
 OperatingSystems 23CDI304

This algorithm suffers from the situation in which a page is used heavily during
theinitial phase of a process but never used again. Since it was used heavily, it has a
largecountandremains inmemoryeventhoughit is no longerneeded.
b) MFUAlgorithm:
based on the argument that the page with the smallest count was probably just brought
inandhasyetto be used

ALLOCATIONOFFRAMES
 The absolute minimum number of frames that a process must be allocated is
dependentonsystem architecture.
 Themaximumnumberisdefinedbytheamountof availablephysicalmemory.
AllocationAlgorithms
After loading of OS, there are two ways in which the allocation of frames can be done
totheprocesses.
Equal Allocation- If there are m frames available and n processes to share them,
eachprocessgetsm/nframes,andtheleftover’sarekeptinafree-framebufferpool.
Proportional Allocation - Allocate the frames proportionally depending on the sizeof
the process. If the size of process i is Si, and S is the sum of size of all processes inthe
system, then the allocation for process Pi is ai= m * Si/ S. where m is the
freeframesavailableinthesystem.
 Consider a system with a 1KB frame size. If a small student process of 10 KB and
aninteractive database of 127 KB are the only two processes running in a system with
62freeframes.
 withproportionalallocation,wewouldsplit62framesbetweentwoprocesses,asfollows
m=62,S=(10+127)=137
Allocationforprocess1=62X10/137~4Allocationforprocess2=62X127/137 ~57
Thusallocates4framesand57framestostudentprocessanddatabaserespectively.
 Variationsonproportionalallocationcouldconsiderpriorityofprocessratherthanjusttheirsiz
e.
GlobalversusLocalAllocation
 Pagereplacementcanoccurbothatlocalorgloballevel.
 With local replacement, the number of pages allocated to a process is fixed, and
pagereplacement occursonlyamongstthepagesallocatedtothisprocess.
 With global replacement, any page may be a potential victim, whetherit
currentlybelongstotheprocessseekingafreeframeornot.
 Local page replacement allows processes tobetter control their own page fault rates,and
leads to more consistent performance of a given process over different system
loadlevels.

32 Dept.of CSE(DS),SJBIT
 OperatingSystems 23CDI304

 Global page replacement is over all more efficient, and is the more commonly
usedapproach.

Non-UniformMemoryAccess (New)
 Usuallythe timerequiredtoaccessall memoryinasystemisequivalent.
 This may not be the case in multiple-processor systems, especially where each CPU
isphysically located on a separate circuit board which also holds some portion of
theoverallsystemmemory.
 In such systems, CPU s can access memory that is physically located on the same
boardmuchfasterthanthememoryontheotherboards.
 Thebasicsolutionis akintoprocessoraffinity -Atthesametimethatwetry toschedule
processes on the same CPU to minimize cache misses, we also try to
allocatememoryforthoseprocessesonthesameboards,tominimizeaccesstimes.

THRASHING
 If the number of frames allocated to a low-priority process falls below the
minimumnumberrequiredbythe
computerarchitecturethenwesuspendtheprocessexecution.
 Aprocessisthrashingifitisspending moretimeinpagingthanexecuting.
 If the processes do not have enough number of frames, it will quickly page
fault.During this it must replace some page that is not currently in use. Consequently
itquicklyfaultsagainandagain.
 The process continuesto fault, replacing pages for whichit thenfaults and bringsback.
This high paging activity is called thrashing. The phenomenon of
excessivelymovingpagesbackandforthb/w
memoryandsecondaryhasbeencalledthrashing.

CauseofThrashing
 Thrashingresultsinsevereperformanceproblem.
 Theoperatingsystemmonitorsthecpuutilizationislow.Weincreasethedegreeofmultiprogra
mmingbyintroducingnewprocesstothesystem.
 Aglobalpagereplacementalgorithmreplacespageswithnoregardstotheprocesstowhichthe
ybelong.

33 Dept.of CSE(DS),SJBIT
 OperatingSystems 23CDI304
Thefigureshowsthethrashing
 Asthedegreeofmultiprogrammingincreases,moreslowlyuntilamaximumis

34 Dept.of CSE(DS),SJBIT
 OperatingSystems 23CDI304

reached.Ifthedegreeofmultiprogrammingisincreasedfurtherthrashingsetsinandthecpuutiliz
ationdropssharply.
 Atthispoint,toincreasesCPUutilizationandstopthrashing,wemustincreasedegreeofmultip
rogramming.
 wecanlimittheeffectofthrashingbyusingalocalreplacementalgorithm.Topreventthrashing
,wemustprovideaprocessasmanyframesasitneeds.

LocalityofReference:
 Astheprocessexecutesit movesfromlocalitytolocality.
 Alocalityisasetofpagesthatareactivelyused.
 Aprogrammayconsistof severaldifferentlocalities,whichmayoverlap.
 Localityiscausedbyloopsincodethatfindtoreferencearraysandotherdatastructuresbyindic
es.
 Theorderedlistofpagenumberaccessedbyaprogramiscalledreferencestring.
 Localityis oftwotypes:
1.spatiallocality 2.temporallocality

Workingsetmodel
 Working set model algorithm uses the current memory requirements to determine
thenumberofpageframestoallocatetotheprocess,aninformaldefinitionis“thecollection of
pages that a process is working with and which must be resident if theprocess to avoid
thrashing”. The idea is to use the recent needs of a process to predictitsfuture reader.
 The working set is an approximation of programs locality. Ex: given a sequence
ofmemory reference, if the working set window size to memory references, then
workingsetattimet1 is{1,2,5,6,7}andatt2ischangedto{3,4}
 At any given time, all pages referenced by a process in its last 4 seconds of
executionareconsidered to compromise itsworking set.
 Aprocess willneverexecuteuntilitsworking setisresidentin main memory.
 Pagesoutsidetheworkingsetcanbediscardedatanymovement.
 Workingsetsarenotenoughandwemustalsointroducebalanceset.
 If the sum of the working sets of all the run able process is greater than the
sizeofmemorytherefusesomeprocessforawhile.
 Divide the run able process into two groups, active and inactive. The
collectionof active set is called the balance set. When a process is made active its
workingsetis loaded.
 Some algorithm must be provided for moving process into and out of the
balanceset. As a working set is changed, corresponding change is made to the
balanceset.
 Working set presents thrashing by keeping the degree of multi programming
ashigh as possible. Thusif optimizes the CPUutilization. Themain
disadvantageofthisiskeepingtrackoftheworkingset.

35 Dept.of CSE(DS),SJBIT
 OperatingSystems 23CDI304

Page-FaultFrequency
 When page- fault rate is too high, the process needs more frames and when it is
too low,theprocessmayhavetoomanyframes.
 Theupperandlowerboundscanbeestablishedonthepage-faultrate.Iftheactualpage-
fault rate exceeds the upper limit, allocate the process another frame or
suspendtheprocess.
 Ifthe page-fault rate falls below the lower limit, remove a framefrom the
process.Thus,wecandirectlymeasureandcontrolthepage-
faultratetopreventthrashing.

36 Dept.of CSE(DS),SJBIT
 OperatingSystems 23CDI304

QUESTIONBANK

DEADLOCKS

1. Whataredeadlocks?Whatareitscharacteristics?Explainthenecessaryconditionsforitsoccurr
ence.
2. Explaintheprocessofrecoveryfromdeadlock.
3. DescribeRAG:
i) Withdeadlock
ii) Withacyclebutnodeadlock
4. WhatisResourceAllocationGraph(RAG)?ExplainhowRAGisveryusefulindescribingdeadl
yembrace(deadlock)byconsideringyourownexample.
5. Withthehelpofasystemmodel,explainadeadlockandexplainthenecessaryconditions
thatmust holdsimultaneouslyinasystemforadeadlocktooccur.
6. Explain how deadlock can be prevented by considering four necessary
conditions cannothold.
7. UsingBanker's algorithmdetermineswhether thesystemisina safestate.
8. Howisasystemrecoveredfromdeadlock?Explainthedifferentmethodsusedtorecoverfromde
adlock.
9. Explaindeadlockdetectionwithalgorithmandexample
10. Definetheterms:safestateandsafesequence.Giveanalgorithmtofindwhetherornotasystemisi
nasafestate.

MEMORYMANAGEMENT

1. Explainthemultistepprocessingof auserprogramwitha neatblockdiagram.

2. Distinguishbetweeninternalandexternalfragmentation.
3. Explainsegmentation withanexample.
4. Explainwithadiagram,howTLB isusedtosolvetheproblemof simplepagingscheme.
5. With a supporting paging hardware, explain in detail concept of paging with an
examplefor a 32-byte memory with 4-type pages with a process being 16-bytes. How
many bitsare reserved for page number and page offset in the logical address. Suppose
the logicaladdress is 5, calculate the corresponding physical address, after populating
memory andpagetable.
6. Whatarethedrawbacksof contiguous memoryallocation?
7. Considerapaging systemwiththepagetable storedinmemory.
i. if a memory reference takes 200 nano seconds, how long does a paged
memoryreferencetake?
ii. if we add associative register and 75 percentage of all page table references
arefound in the associative registers, what is the effective memory access time?
(Assumethatfindingapagetableentryintheassociativememory/registerstakes

37 Dept.of CSE(DS),SJBIT
OperatingSystems 23CDI304

zerotime,iftheentryisfound).
8. Distinguishbetween:
i. Logicaladdressspace andphysicaladdressspace.
ii. Internalfragmentation andexternalfragmentation.
iii. Pagingand segmentation.
9. ExplainwiththehelpofsupportinghardwarediagramhowtheTLBimprovestheperformanceof
ademandpagingsystem.
10. Explainthe concept of forwardmappedpagetable.
11. Whatisfragmentation?Explaintwotypesof memoryfragmentation
12. Whatisswapping?Explainindetail.
13. Whatdoyoumeanbyaddressbinding?Explainwiththenecessarysteps,thebindingofinstructio
nsanddatatomemoryaddresses.

38 Dept.of CSE(DS),SJBIT
 OperatingSystems 23CDI304
UniversityQuestions

June/July2018

Dec.2018/ Jan2019

39 Dept.of CSE(DS),SJBIT
 OperatingSystems 23CDI304
June/July2019

40 Dept.of CSE(DS),SJBIT