1. Explain the concept of Reentrancy?

It is a useful, memory-saving technique for multiprogrammed timesharing systems. A Reentrant Procedure is

one in which multiple users can share a single copy of a program during the same period. Reentrancy has 2
key aspects !he program code cannot modify itself, and the local data for each user process must "e stored
separately. !hus, the permanent part is the code, and the temporary part is the pointer "ack to the calling
program and local varia"les used "y that program. #ach e$ecution instance is called activation. It e$ecutes
the code in the permanent part, "ut has its own copy of local varia"les%parameters. !he temporary part
associated with each activation is the activation record. &enerally, the activation record is kept on the stack.
Note A reentrant procedure can "e interrupted and called "y an interrupting program, and still e$ecute
correctly on returning to the procedure.
2. Explain Belady's Anomaly?
Also called 'I'( anomaly. )sually, on increasing the num"er of frames allocated to a process virtual memory,
the process e$ecution is faster, "ecause fewer page faults occur. *ometimes, the reverse happens, i.e., the
e$ecution time increases even when more frames are allocated to the process. !his is +elady,s Anomaly. !his
is true for certain page reference patterns.
3. What is a binary semaphore? What is its se?
A "inary semaphore is one, which takes only - and . as values. !hey are used to implement mutual
e$clusion and synchroni/e concurrent processes.
!. What is thrashin"?
It is a phenomenon in virtual memory schemes when the processor spends most of its time swapping pages,
rather than e$ecuting instructions. !his is due to an inordinate num"er of page faults.
#. $ist the %offman's conditions that lead to a deadloc&.
.. 'tal Exclsion (nly one process may use a critical resource at a time.
2. (old ) Wait A process may "e allocated some resources while waiting for others.
0. No *re+emption 1o resource can "e forci"le removed from a process holding it.
2. %irclar Wait A closed chain of processes e$ist such that each process holds at least one resource
needed "y another process in the chain.
,. What are short- lon" and medim+term schedlin"?
$on" term schedler determines which programs are admitted to the system for processing. It controls the
degree of multiprogramming. (nce admitted, a 3o" "ecomes a process.
'edim term schedlin" is part of the swapping function. !his relates to processes that are in a "locked or
suspended state. !hey are swapped out of real-memory until they are ready to e$ecute. !he swapping-in
decision is "ased on memory-management criteria.
.hort term schedler, also know as a dispatcher e$ecutes most frequently, and makes the finest-grained
decision of which process should e$ecute ne$t. !his scheduler is invoked whenever an event occurs. It may
lead to interruption of one process "y preemption.
/. What are trnarond time and response time?
!urnaround time is the interval "etween the su"mission of a 3o" and its completion. Response time is the
interval "etween su"mission of a request, and the first response to that request.
0. What are the typical elements of a process ima"e?
1ser data 4odifia"le part of user space. 4ay include program data, user stack area, and programs that
may "e modified.
1ser pro"ram !he instructions to "e e$ecuted.
.ystem .tac& #ach process has one or more 5I'( stacks associated with it. )sed to store parameters and
calling addresses for procedure and system calls.
*rocess control Bloc& 6P7+8 Info needed "y the (* to control processes.
2. What is the 3ranslation $oo&aside Bffer 43$B5?
In a cached system, the "ase addresses of the last few referenced pages is maintained in registers called the
!5+ that aids in faster lookup. !5+ contains those page-ta"le entries that have "een most recently used.
1ormally, each virtual memory reference causes 2 physical memory accesses- one to fetch appropriate page-
ta"le entry, and one to fetch the desired data. )sing !5+ in-"etween, this is reduced to 3ust one physical
memory access in cases of !5+-hit.
16. What is the resident set and 7or&in" set of a process?
Resident set is that portion of the process image that is actually in real-memory at a particular instant.
9orking set is that su"set of resident set that is actually needed for e$ecution. 6Relate this to the varia"le-
window si/e method for swapping techniques.8
11. When is a system in safe state?
!he set of dispatcha"le processes is in a safe state if there e$ists at least one temporal order in which all
processes can "e run to completion without resulting in a deadlock.
12. What is cycle stealin"?
9e encounter cycle stealing in the conte$t of :irect 4emory Access 6:4A8. #ither the :4A controller can use
the data "us when the 7P) does not need it, or it may force the 7P) to temporarily suspend operation. !he
latter technique is called cycle stealing. 1ote that cycle stealing can "e done only at specific "reak points in
an instruction cycle.
13. What is meant by arm+stic&iness?
If one or a few processes have a high access rate to data on one track of a storage disk, then they may
monopoli/e the device "y repeated requests to that track. !his generally happens with most common device
scheduling algorithms 65I'(, **!', 7-*7A1, etc8. ;igh-density multisurface disks are more likely to "e
affected "y this than low density ones.
1!. What are the stiplations of %2 le8el secrity?
72 level security provides for
.. :iscretionary Access 7ontrol
2. Identification and Authentication
0. Auditing
2. Resource reuse
1#. What is bsy 7aitin"?
!he repeated e$ecution of a loop of code while waiting for an event to occur is called "usy-waiting. !he 7P)
is not engaged in any real productive activity during this period, and the process does not progress toward
completion.
1,. Explain the poplar mltiprocessor thread+schedlin" strate"ies.
.. $oad .harin" Processes are not assigned to a particular processor. A glo"al queue of threads is
maintained. #ach processor, when idle, selects a thread from this queue. 1ote that load "alancing
refers to a scheme where work is allocated to processors on a more permanent "asis.
2. 9an" .chedlin" A set of related threads is scheduled to run on a set of processors at the same
time, on a .-to-. "asis. 7losely related threads % processes may "e scheduled this way to reduce
synchroni/ation "locking, and minimi/e process switching. &roup scheduling predated this strategy.
0. :edicated processor assi"nment Provides implicit scheduling defined "y assignment of threads
to processors. 'or the duration of program e$ecution, each program is allocated a set of processors
equal in num"er to the num"er of threads in the program. Processors are chosen from the availa"le
pool.
2. :ynamic schedlin" !he num"er of thread in a program can "e altered during the course of
e$ecution.
1/. When does the condition 'rende;8os' arise?
In message passing, it is the condition in which, "oth, the sender and receiver are "locked until the message
is delivered.
10. What is a trap and trapdoor?
!rapdoor is a secret undocumented entry point into a program used to grant access without normal methods
of access authentication. A trap is a software interrupt, usually the result of an error condition.
12. What are local and "lobal pa"e replacements?
5ocal replacement means that an incoming page is "rought in only to the relevant process address space.
&lo"al replacement policy allows any page frame from any process to "e replaced. !he latter is applica"le to
varia"le partitions model only.
26. :efine latency- transfer and see& time 7ith respect to dis& <=>.
*eek time is the time required to move the disk arm to the required track. Rotational delay or latency is the
time it takes for the "eginning of the required sector to reach the head. *um of seek time 6if any8 and
latency is the access time. !ime taken to actually transfer a span of data is transfer time.
21. :escribe the Bddy system of memory allocation.
'ree memory is maintained in linked lists, each of equal si/ed "locks. Any such "lock is of si/e 2<k. 9hen
some memory is required "y a process, the "lock si/e of ne$t higher order is chosen, and "roken into two.
1ote that the two such pieces differ in address only in their kth "it. *uch pieces are called "uddies. 9hen any
used "lock is freed, the (* checks to see if its "uddy is also free. If so, it is re3oined, and put into the original
free-"lock linked-list.
22. What is time+stampin"?
It is a technique proposed "y 5amport, used to order events in a distri"uted system without the use of
clocks. !his scheme is intended to order events consisting of the transmission of messages. #ach system ,i, in
the network maintains a counter 7i. #very time a system transmits a message, it increments its counter "y .
and attaches the time-stamp !i to the message. 9hen a message is received, the receiving system ,3, sets its
counter 73 to . more than the ma$imum of its current value and the incoming time-stamp !i. At each site,
the ordering of messages is determined "y the following rules 'or messages $ from site i and y from site 3, $
precedes y if one of the following conditions holds....6a8 if !i=!3 or 6"8 if !i>!3 and i=3.
23. (o7 are the 7ait=si"nal operations for monitor different from those for semaphores?
If a process in a monitor signal and no task is waiting on the condition varia"le, the signal is lost. *o this
allows easier program design. 9hereas in semaphores, every operation affects the value of the semaphore,
so the wait and signal operations should "e perfectly "alanced in the program.
2!. <n the context of memory mana"ement- 7hat are placement and replacement al"orithms?
Placement algorithms determine where in availa"le real-memory to load a program. 7ommon methods are
first-fit, ne$t-fit, "est-fit. Replacement algorithms are used when memory is full, and one process 6or part of
a process8 needs to "e swapped out to accommodate a new program. !he replacement algorithm determines
which are the partitions to "e swapped out.
2#. <n loadin" pro"rams into memory- 7hat is the difference bet7een load+time dynamic lin&in"
and rn+time dynamic lin&in"?
'or load+time dynamic lin&in" 5oad module to "e loaded is read into memory. Any reference to a target
e$ternal module causes that module to "e loaded and the references are updated to a relative address from
the start "ase address of the application module.
9ith rn+time dynamic loadin" *ome of the linking is postponed until actual reference during e$ecution.
!hen the correct module is loaded and linked.
2,. What are demand+pa"in" and pre+pa"in"?
9ith demand paging, a page is "rought into memory only when a location on that page is actually referenced
during e$ecution. 9ith pre-paging, pages other than the one demanded "y a page fault are "rought in. !he
selection of such pages is done "ased on common access patterns, especially for secondary memory devices.
2/. *a"in" a memory mana"ement fnction- 7hile mltipro"rammin" a processor mana"ement
fnction- are the t7o interdependent?
?es.
20. What is pa"e cannibali;in"?
Page swapping or page replacements are called page canni"ali/ing.
22. What has tri""ered the need for mltitas&in" in *%s?
.. Increased speed and memory capacity of microprocessors together with the support fir virtual
memory and
2. &rowth of client server computing
36. What are the for layers that Windo7s N3 ha8e in order to achie8e independence?
.. ;ardware a"straction layer
2. @ernel
0. *u"systems
2. *ystem *ervices.
31. What is .'*?
!o achieve ma$imum efficiency and relia"ility a mode of operation known as symmetric multiprocessing is
used. In essence, with *4P any process or threads can "e assigned to any processor.
32. What are the &ey ob?ect oriented concepts sed by Windo7s N3?
#ncapsulation, ("3ect class and instance.
33. <s Windo7s N3 a fll blo7n ob?ect oriented operatin" system? 9i8e reasons.
1o 9indows 1! is not so, "ecause its not implemented in o"3ect oriented language and the data structures
reside within one e$ecutive component and are not represented as o"3ects and it does not support o"3ect
oriented capa"ilities.
3!. What is a dra7bac& of '@3?
It does not have the features like
.. a"ility to support multiple processors
2. virtual storage
0. source level de"ugging
3#. What is process spa7nin"?
9hen the (* at the e$plicit request of another process creates a process, this action is called process
spawning.
3,. (o7 many ?obs can be rn concrrently on '@3?
.A 3o"s.
3/. $ist ot some reasons for process termination.
.. 1ormal completion
2. !ime limit e$ceeded
0. 4emory unavaila"le
2. +ounds violation
A. Protection error
B. Arithmetic error
C. !ime overrun
D. I%( failure
E. Invalid instruction
.-.Privileged instruction
...:ata misuse
.2.(perator or (* intervention
.0.Parent termination.
30. What are the reasons for process sspension?
.. swapping
2. interactive user request
0. timing
2. parent process request
32. What is process mi"ration?
It is the transfer of sufficient amount of the state of process from one machine to the target machine.
!6. What is mtant?
In 9indows 1! a mutant provides kernel mode or user mode mutual e$clusion with the notion of ownership.
!1. What is an idle thread?
!he special thread a dispatcher will e$ecute when no ready thread is found.
!2. What is At:is&?
It is a fault tolerance disk driver for 9indows 1!.
!3. What are the possible threads a thread can ha8e?
.. Ready
2. *tand"y
0. Running
2. 9aiting
A. !ransition
B. !erminated
!!. What are rin"s in Windo7s N3?
9indows 1! uses protection mechanism called rings provides "y the process to implement separation
"etween the user mode and kernel mode.
!#. What is Execti8e in Windo7s N3?
In 9indows 1!, e$ecutive refers to the operating system code that runs in kernel mode.
!,. What are the sb+components of <=> mana"er in Windo7s N3?
.. 1etwork redirector% *erver
2. 7ache manager.
0. 'ile systems
2. 1etwork driver
A. :evice driver
!/. What are ::&s? Name an operatin" system that incldes this featre.
::ks are device driver kits, which are equivalent to *:@s for writing device drivers. 9indows 1! includes
::ks.
!0. What le8el of secrity does Windo7s N3 meets?
72 level security.