Quiz
Quiz 2
Create a free account to finish this quiz
Save your progress and see your final score.
Already have an account? Sign in
Questions in this quiz (200)
1In some systems with priority-based processes, a problem can occur that results in the highest priority ready process being prevented from running by a lower priority process. This problem is called
- Athe priority inversion problem
- Bthe priority reversal problem
2Suppose a low-priority process L holds a resource needed for the execution of a high-priority process H, but that L is prevented from running because a medium-priority process M is continually ready to run, and only a single processor is available. What is the name of an algorithm that could be used to deal with this problem?
- Aaging
- Breverse sorting
3One technique that can be used to eliminating the busy waiting used with some solutions to the mutual exclusion problem is
- Ato have a process relinquish the processor and block until another process, leaving its critical section, awakens it so it can repeat the test
- Bto disable interrupts during the wait loop
4The processes involved in the producer-consumer problem are those that
- Aoccasionally generate objects and those that then use those objects
- Bare compute-bound and I/O-bound
5The objects that are being produced in the producer-consumer problem are generally characterized as
- Adata objects, like indications of keys pressed or released on a keyboard.
- Bblocks of memory, independent of their content
6The processes in the producer-consumer problem are assumed to run
- Aat different and arbitrary rates, as is required by good solutions to race condition problems
- Bat about the same rate
7A buffer, or storage area, is required in the producer-consumer problem. This buffer holds
- Aobjects produced before a consumer process is ready to use them
- Bobjects that a consumer process considered, but then rejected as unsuitable
8Access to the buffer in the producer-consumer problem
- Ais shared by producers and consumers, and so must occur only inside a critical section
- Bis controlled by the last object removed from the buffer
9When the buffer in the producer-consumer problem is completely full
- Aa producer that has created an object must block until an empty space becomes available in the buffer.
- Bthe number of producer processes is reduced by half.
10Assume a producer process is blocked while waiting on empty space in the buffer. In this case, a consumer process
- Awill awaken a producer when it removes an object from the buffer, allowing space for the producer to place another object
- Bwill remove and discard sufficient data objects (from the buffer) to make space available for the blocked producer to place additional objects in the buffer.
11Consider that the statement count = count + 1; appears in multiple processes or threads in a system where the variable named count is accessible by multiple processes or threads. What must be done to ensure the value of count is correctly incremented when the statement is executed?
- AIt is impossible to guarantee that count will be properly incremented if multiple threads or processes contain the statement that can increment it.
- BThe statement must be in a critical section in each of the processes or threads that contains it.
12Who proposed the semaphore data structure?
- AEdgser Dijkstra
- BDonald Shell
13What is the significance of the letters P and V when describing the fundamental operations on a semaphore?
- AThey correspond to the first letters of the Dutch words that describe the operations.
- BThey correspond to the first letters of the words Permit and Validate, which describe the actions of the fundamental operations.
14The problem with the producer consumer problem solution using the sleep and wakeup functions is related to the fact that
- Awakeup and sleep are not atomic operations.
- Ba wakeup of a process (either a producer or a consumer) could be lost
15A semaphore has two data components. One of these is a set. What type of data objects can this set contain?
- Aprocess identifications
- Bprocessor affinities
16A semaphore has two data components. One of these is an integer. What does the value of this integer represent?
- Athe number of waiting wakeups for processes that request use of a resource controlled by the semaphore
- Bthe number of processes currently holding (possessing) one or more units of the resourcecontrolled by the semaphore
17A P operation on a semaphore is also called a(n) _____ operation
- Adown
- Bleft
18A V operation on a semaphore is also called a(n) _____ operation
- Aup
- Bdown
19When a P operation is executed on a semaphore with a count of zero
- Athe identity of the process is added to the set associated with the semaphore, but the process continues execution.
- Bthe process executing the P operation is blocked
20When a V operation is executed on a semaphore with a count of zero,
- Aif any processes are identified by the set associated with the semaphore, one of them is removed from the set and moved from the blocked to the ready state; otherwise the count is incremented by 1
- Bthe count associated with the semaphore is increased by 1.
21Which of the semaphore operations, P or V, must be executed atomically?
- ABoth operations must be executed atomically.
- BP must be executed atomically
22 In the semaphore-based solution to the producer/consumer problem, both the producer and the consumer processes use access to the shared buffer as a resource, and use it in a mutually-exclusive manner by performing down and up operations on a semaphore named mutex. What other resource do producer processes require if they are not going to be blocked?
- Aempty locations in the buffer into which a producer can place an item
- Baccess to the variable containing a consumer's process identification
23Who proposed the synchronization structure called a monitor?
- ANiklaus Wirth and David Gries
- BTony Hoare and Per Brinch Hansen
24A monitor is different from a semaphore or an event counter in several ways. A semaphore and an event counter are each data structures, but a monitor is
- Aa programming construct intended for inclusion in a programming language.
- Ba set of processes, each of which has access to a single resource at all times.
25 It is the responsibility of the compiler that processes the programming language supporting monitors to guarantee that
- Aat most one process or thread is sleeping on any of the monitor's condition variables at any time.
- Bonly one process (or thread) is allowed to execute any of the functions or the initialization code inside the monitor at one time.
26A condition variable used with a monitor is similar to a semaphore in that it includes a set. It is different from a semaphore, however, in what way?
- AA condition variable includes two counts, one giving the number of resources in use, and one giving the number of processes waiting for a resource.
- BThere is no count associated with a condition variable
27 A signal operation on a condition variable associated with a monitor, with no processes waiting on the condition variable,
- Ais ignored.
- Bresults in an error.
28Brinch Hansen and Hoare each proposed a different way of dealing with a process P that signaled a condition variable and caused another process, say process Q, to be awakened. Why was it necessary to devise such strategies?
- AThe signaling process, P, is not allowed to continue execution inside the monitor while the awakened process, Q, is also executing inside the monitor.
- BThe semantics associated with the signal operation were not well known at the time it was proposed, and so additional work was required to refine the semantics.
29 Which of the following synchronization techniques can easily be used to support processes running on separate machines connected to a network?
- Amessage passing
- Bevent counters
30When a process executes the appropriate code to receive a message, what will likely happen if a message is not immediately available for it to receive?
- AThe process will block until such time as a message is available.
- BAn error will be reported and the process attempting to receive the message is terminated.
31 In the producer-consumer solution using message passing, there is some initialization required. What takes place during this one time only initialization?
- AAn appropriate number of messages, each marked as being empty, are sent from the consumer to the producer.
- BEach of the consumer processes is placed in the blocked state.
32Which of the following statements about the equivalence of process synchronization techniques is true?
- AEach of the techniques - monitors, event counters, semaphores, and message passing -can be shown to be equivalent to each of the others.
- BMonitors and event counters can be shown to be equivalent
33In the dining philosophers problem, the philosophers spend their lives alternating between thinking and
- Aworking
- Beating
34The shared resource(s) in the dining philosophers problem is(are)
- Afood
- Bforks.
35The processes in the dining philosophers problem correspond to the behavior of
- Athe food supply
- Bthe philosophers
36Suppose there are 7 philosophers in the dining philosophers problem. At most how many of these could be eating at the same time?
- A2
- B3
37Suppose there are 8 philosophers in the dining philosophers problem. What is the minimumnumber of these that may be eating at the same time?
- A5
- B0
38When all that is impossible is eliminated, then whatever remains, however improbable,
- Amust be the truth.
- Bmust be eliminated.
39One way to find potential problems with a proposed solution to an IPC problem is to
- Aconsider all possible execution sequences and eliminate those that are not causing problems.
- Brun the proposed solution on a system with one processor for each process
40One sequence of actions that must not be allowed in the dining philosophers problem is
- Afor each philosopher to pick up the fork to their left and then, while holding it, wait for the fork to their right
- Bfor all philosophers to be thinking at the same time
41A good solution to the dining philosophers problem must provide a reasonable measure of assurance that
- Aonce a philosopher has indicated a desire to eat that he or she will be eventually allowed to do so
- Bno more than three other philosophers will be allowed to eat before a philosopher that is hungry will be allowed to eat
42Suppose in a proposed solution to the dining philosopher problem that a philosopher has indicated it is hungry, but it is forever denied the opportunity to eat. In this case, we say that the solution exhibits
- Aindefinite postponement
- Btemporal instability
43In the solution to the dining philosophers problem presented, a third state is associated with each philosopher. This additional state is called
- Ahungry
- Bright fork
44How many classes of processes are there in the readers/writers problem?
- Amore than 4
- B2
45Each process in the readers/writers problem is involved with reading or writing what?
- Aa shared object, like a database
- Bone or more objects, none of which may be accessed by more than one reader at a time
46How many reader processes can be accessing an object in the readers/writers problem at the same time?
- Aas many as there are writer processes
- Bas many as desired
47How many writer processes can be accessing an object in the readers/writers problem at the same time?
- Aat most one
- Bat most two
48In total, how many reader and writer processes may be accessing an object in the readers/writers problem at the same time?
- AAn equal number of readers and writers may be accessing an object at the same time.
- BReaders and writers may not both access the shared object at the same time
49Which of the following are common variants of the reader/writer problem?
- Areader priority and writer priority
- Bstatic writer and dynamic writer
50Assume we allow reader processes to continually gain access to the shared object in the reader/writer problem, even if writers are waiting to access it. Which of the following is true?
- AIndefinite postponement of writers could occur.
- BIndefinite postponement of readers and writers could occur.
51The solution presented for the reader/writer problem uses two semaphores. What resources have their access controlled by these semaphores?
- Athe variable that indicates the number of active reader processes, and the variable that indicates the number of active writer processes
- Bthe shared object and the variable that indicates which type of processes are currently accessing the shared object
52Assume there is a semaphore DB used to control access to the shared object in the reader/writer problem. Also assume there are four readers currently accessing the shared object. How many of those readers did a down operation on the DB semaphore to obtain their current access to the shared object?
- Aone or none
- Bthree
53Assume there is a semaphore used to control access to the shared object in the reader/writer problem. Also assume five write operations have been performed on the shared object. How many successful down operations were performed by writer processes on the semaphore?
- Aat least two but no more than five
- Bfive
54The term batch processing historically refers to
- Athe processing of a set of jobs that were all submitted as a group.
- Bthe processing of requests from a batch of users at terminals.
55Batch processing today is most likely referred to as
- Aembedded processing
- Bbackground processing
56Jobs processed in batches
- Amust all have been submitted from the same interactive terminal
- Busually do not perform I/O on terminals or other interactive devices
57Interactive jobs
- Acan only perform input/output using the local keyboard, display, and mouse.
- Butilize terminals, mice, and other devices designed for human input and output
58Interactive processes
- Atypically require timely response from the system after user input occurs.
- Bmust always be started from the terminal directly connected to the system, never through a network terminal.
59The term scheduling refers to
- Aall activities of a system associated with determining when various actions are to be performed.
- Bthe processing of requests from processes in an order that is strictly the same as the order in which the requests were submitted.
60If a typical process is executed several times with the same data
- Ait should execute exactly the same sequence of instructions
- Bit will produce output that is dependent on the number of times the same code is beingexecuted; that is, how many processes are executing the same text editor, for example
61Regardless of the number of processes being executed, or the code that processes are executing
- Aeach possible execution sequence of each process is finite
- Beach process will receive the same amount of CPU time
62The term job scheduling refers to
- Athe determination of when individual jobs are submitted to a pool, or queue, of jobs from which individual jobs may be selected for execution by the CPU.
- Bthe scheduling of a job, or sequence of individual sequential steps
63Job scheduling is done using
- Athe number of times the job has been run over the past several time intervals in which scheduling decisions are made.
- Binformation about the resources required by the entire job, and the currently available system resources.
64Which of the following would be characterized as high-level scheduling?
- Ajob scheduling
- Binstruction scheduling
65Which of the following would be characterized as low-level scheduling?
- Ajob scheduling
- Bprocess scheduling
66to what does the term turnaround time apply?
- Athe time required to switch a processor (CPU) from one process to another.
- Bthe time between submitting a non-interactive, or batch, job and receiving the results
67Which of the following is not a goal of scheduling?
- Agive interactive users timely responses to their actions.
- Bkeep the CPU as idle as possible, reflecting effective use of the processor
68When preemptive scheduling is used
- Aa process can continue to use the processor as long as it wishes, or is capable of using it.
- Ba running process can have the CPU taken away from it, and it can be returned to the ready state
69The term non-preemptive scheduling is also called
- Ainteractive scheduling
- Brun-to-completion scheduling.
70A clock in a computer system operates by
- Aproviding periodic interrupts
- Bmeasuring the rate at which the CPU is executing processes.
71Which of the following actions might be taken each time the primary system clock ticks?
- AIf any actions are scheduled to occur at the current time, then steps are taken to start those actions
- BEvery process in the blocked state is checked to see if the resource (or resources) forwhich it is waiting is now available
72Suppose there are only three processes in a system, and they are compute-bound (that is,they do no input/output at all). Assume the processes are using round-robin scheduling with a quantum size of 10 milliseconds. Suppose each process needs 1 second (1000 milliseconds) of CPU time to complete its work. Ignoring the time required to switch theCPU between processes, after how long will the first process complete its work?
- A2990 milliseconds
- B2980 milliseconds
73The maximum amount of time a process may use the CPU before it is returned to the ready state in a round-robin system is called
- Athe quantum size of the process.
- Bthe recycle period of the system
74If the quantum size in a system is very small, then most of the time in the system is spent
- Aperforming context switches between processes
- Bexecuting the idle process
75If the quantum size in a system is very large
- Amemory utilization will be very efficient
- Bthe system effectively appears to be doing non-preemptive scheduling.
76In a system that uses priority scheduling
- Athe process currently using the CPU is one of those that has the highest priority
- Bit is not also possible to use round-robin scheduling.
77In a system that uses priority scheduling
- Aprocesses with the same priority are allowed only if the system has multiple processors.
- Bprocesses with the same priority can be run in a round-robin fashion
78In a UNIX system, when a process blocks for an input/output operation
- Athe system assumes it is I/O bound (that is, spending most of its time performing input/output), and its priority is increased
- Bits priority is changed to the lowest of all ready processes.
79Suppose a system is using the inverse remainder of quantum variant of round-robin scheduling. A process uses 30 percent of its quantum before blocking. When it is eventually returned to the ready queue,
- Aits priority is increased by 30 percent
- Bit is placed behind 30 percent of the processes in the ready queue, not at the rear
80The batch scheduling algorithm called shortest job first always selects the job with the smallest execution time to run next. This has the result of
- Aalways yielding the smallest average turnaround time of any job scheduling algorithm.
- Byielding the smallest standard deviation of the job turnaround times of any job scheduling algorithm
81Since the shortest job first algorithm cannot be used directly on interactive processes, a variation called aging is used. In this algorithm
- Athe estimated execution time for each process is increased by a fixed percentage each time the process enters the running state.
- Bthe estimated execution time for a process is a weighted sum of the previous execution time estimate and the last execution time.
82For many systems, the only criteria associated with the correct execution of a process iswhether the process yields the correct results. Systems in which it is also required thatactions be taken in a specified amount of time or at specified times are called
- Apriority-based systems
- Breal-time systems
83A system in which failing to meet a deadline can result in total system failure or death is
- Acalled a hard real-time system
- Bpermitted only if multiple processors are available
84Suppose a real-time system is going to process only two types of events. The first type of event occurs once every 10 milliseconds and requires 4 milliseconds of CPU time to process.The second event occurs once every 20 milliseconds. What is the most CPU time each2 0-millisecond event can require if the system is to be considered schedulable?
- A8 milliseconds
- B12 milliseconds
85In the rate-monotonic scheduling algorithm
- Aprocesses are run to completion once they are selected for execution.
- Beach process gets a priority that is proportional to the frequency of occurrence of its triggering event
86The term swapping describes a system that
- Amoves the memory content of processes between primary memory and secondary storage if insufficient memory is available.
- Bgives each process two priorities which are swapped when the process moves between the ready/running states and the blocked state.
87Which of the following statements regarding the terms policy and mechanism is correct?
- AA policy is a set of rules, or parameters associated with a set of rules, that are implemented using the available mechanisms in a system.
- BPolicies are tools available in an operating system to implement various mechanisms that may be desired
88Deadlock occurs when each process in a set of processes
- Ais waiting for an event (or exclusive-use resource) that only another process can cause(or release).
- Bterminates without closing all of the files it had open.
89Which of the following resources is not one which can result in deadlocking processes?
- Aa hardware interrupt
- Ba disk file
90Which of the following is an example of a consumable resource?
- Aan I/O device
- Ba message sent through some sort of channel
91Which of the following is an example of a non-consumable resource?
- Aa memory region
- Ba message sent through some sort of channel
92Which of the following resources is considered preemptible by many operating systems?
- Aprinters
- Bprimary memory
93What usually happens to a process if a resource it is using is preempted?
- AThe process is notified that the resource has been preempted, and is given an opportunity to decide how to continue
- BThe process is blocked until the resource is restored
94What is the state of a resource that is restored to a process from which it was previously preempted?
- AThe resource will have the state left by the process that used it during the time it was preempted
- BThe resource will have exactly the same state as when it was previously preempted.
95Suppose a process requests a shared lock on a resource, and another process already holds a shared lock. What will happen?
- Athe lock will be converted to an exclusive lock, and the resource will be preempted from the process that held the shared lock
- Bthe process will be granted the shared lock
96There are four necessary conditions required for deadlock. Who first demonstrated these conditions?
- ACoffman
- BBrinch Hansen
97Which of the following is not a necessary condition for deadlock?
- Acircular wait
- Bat least three processes
98What is meant by the hold and wait condition?
- AProcesses must hold exclusive use of a semaphore if they wish to wait for another resource.
- BProcesses are allowed to continue holding (for exclusive use) resources while waiting (blocking) for additional resources.
99What is meant by the no preemption condition?
- AAll of the resources must be preemptible.
- BNone of the resources held by processes can be preempted from those processes
100A resource graph has two types of nodes. What do these nodes represent?
- Aresources and processes
- Bconsumable and non-consumable resources
101Assume we have a resource graph that represents a deadlocked state. Also assume we start from a node associated with a deadlocked process and follow directed edges. Which of the following nodes can we always reach in this edge-following?
- Aa node associated with a preemptible resource
- Bthe node from which we started
102Consider a given resource graph. Where can we find the nodes for the processes that are deadlocked?
- AThey are the process nodes in the cycle (or cycles) in the graph, if any.
- BThey include each process node with at least one edge pointing to it and at least one edge pointing away from it.
103What characterizes the algorithm Tanenbaum calls the ostrich algorithm?
- AThe algorithm searches for process nodes with two incoming edges and one outgoing edge.
- BIt basically means we do nothing, ignoring the possibility of deadlock
104Why might the ostrich algorithm be chosen for a system?
- AThe system was designed to be deadlock free, or the probability of deadlock is so small as to not justify the overhead of one of the other algorithms to deal with deadlock
- BAlthough expensive to implement, the ostrich algorithm is guarantee to prevent deadlock
105The detect and recover approach to dealing with deadlock obviously requires that we detect when deadlock has occurred, and then recover from it. How may the recovery from deadlock be performed?
- ATerminate processes in the reverse order they were started, terminating the last started process first, until the deadlock disappears.
- BIdentify a deadlocked process and then terminate it, releasing the resources to which it had exclusive use
106The detect and recover approach to dealing with deadlock obviously requires that we detect when deadlock has occurred, and then recover from it. Which of the following approaches may be used to detect deadlocked processes without constructing a resource graph and checking for cycles?
- AFind processes that have remained blocked for a considerable period of time.
- BFind processes that have not produced any output for a considerable period of time.
107What is done when the technique called checkpointing is used?
- AAll information about a running process is periodically recorded and saved so it can be used to restore the process from the saved checkpoint later if desired
- BThe resource allocation graph is periodically saved so changes in the graph can be detected.
108What is the essence of techniques used to prevent deadlock?
- AThey deny, or prevent, the occurrence of one (or more) of the necessary conditions for deadlock.
- BThey restrict a process from acquiring exclusive use of any additional resources until it releases exclusive use of all the resources it currently holds.
109To eliminate the need for exclusive use of a resource
- Athe resource must be replicated a sufficient number of times, one per process
- Btechniques must be used to allow the resource to be shared
110What is the traditional approach to allowing a single printer to be shared by many processes?
- AEach process writes its printed output to a file; when the output is complete, the file is added to a queue of files to be printed by a single process that has exclusive use of the printer.
- BEach process gets exclusive use of the printer to print a single line of output at a time.
111How can the hold and wait condition, required for deadlock, be reasonably eliminated?
- AWe can require processes to request all resources they require at one time, and not allow them to use any resources until all the resources are available
- BRequire processes to continually poll for exclusive use of the required resources.
112Which of the four conditions necessary for deadlock does Havender's standard allocation pattern prevent?
- Acircular wait
- Bhold and wait
113Suppose three resource types, A, B, and C are assigned numeric codes 1, 2, and 3 for use with Havender's standard allocation pattern. A process has requested, and been given, exclusive use of resource B. Which of the following actions is then not permitted?
- Arequesting use of resource A
- Brequesting use of an additional unit of resource B
114Deadlock prevention and deadlock avoidance are separate approaches to deal with the potential for deadlock in a system. Which of these approaches eliminates certain system states that might not lead to deadlock?
- Adeadlock prevention
- Bdeadlock avoidance
115Each time a process acquires or releases exclusive use of a resource, the system enters a new state as indicated by a point on a resource trajectory. What characterizes a state that is called unsafe?
- AThe system is not deadlocked, but depending on which process is next allowed to allocate a resource, the system could become deadlocked
- BA process has requested more units of a resource than are available in the system
116Who is responsible for the Banker's algorithm?
- AWirth
- BDijkstra
117What does the Banker's algorithm do?
- AAfter granting each resource request, it determines if the system is in an unsafe state, and if it is, the process state is restored to that specified in its last checkpoint
- BPrior to granting each resource request, it determines if the grant would put the system in an unsafe state, and if it would, the request is deferred
118Why is the Banker's algorithm named as it is?
- AIt suggests the actions taken by a banker with a limited amount of funds to loan, and a number of clients who may collectively want to borrow more than the banker has.
- BIt was developed by Jonas Banker
119In two-phase locking, a process sequentially attempts to exclusively allocate the needed resources. If one of the needed resources is not available, the process
- Arelinquishes all of the resources that were successfully allocated, delays a short (perhaps random) period of time, then repeats the request
- Brelinquishes the last resource requested, waits a random time, and then tries again.
120A block input/output device has which of the following characteristics?
- AIt transfers data in blocks that are relatively large
- BIt is always used to communicate directly with a user
121A device driver
- Ais hardware (i.e. electronics) that stores data until a device is capable of processing it
- Bis software that handles the detailed interaction between the system and a device
122A system has two floppy disk drives. How many floppy disk device drivers are probably present in the system?
- Aone, since similar or identical devices can usually share the same device driver
- Bat least two, but perhaps more than two, since the potential exists for additional drives to be added to the system
123What is the relationship between a device driver and a device controller?
- AA device controller is the software that communicates with a device driver.
- BA device driver is the software that communicates with a device controller.
124The interface between a device driver and the rest of the operating system
- Ais normalized, so every device driver presents the same interface to the operating system
- Bdepends on whether the device is a storage device or not
125Which of the following is not a standard bus acronym?
- APCI-X
- BPCB
126What is a data channel?
- Aa secondary bus on a mainframe computer system that provides reliability in case the primary bus fails
- Ba separate processor on a mainframe that directs the flow of data between I/O devices and primary memory
127The collection of all tracks that can be read or written with a single setting of the heads on a disk is called
- Aa school
- Ba cylinder
128The time required to move the read/write heads on a disk from their current position to a new position is called
- Athe movement delay
- Bthe seek time
129The time required for the desired sector to move so it is appropriately located under a read/write head is called
- Athe transfer time
- Bthe rotational latency
130The time required for data to move between a disk drive and primary memory, once the heads and the disk have been put into the appropriate position, is called
- Athe block time
- Bthe transfer time
131The time required to complete a disk I/O operation is the sum of
- Athe seek time, the rotational latency, and the transfer time.
- Bthe movement delay, the head latency, and the seek time.
132When a disk is not being accessed
- Athe disk rotation may be stopped to save power
- Bthe heads are moved to cylinder 0 in preparation for the next read/write operation
133The smallest unit of data which may be read or written by a physical disk drive is called
- Aa sector
- Ba cylinder
134The set of tracks on a single disk surface
- Ais arranged (at least conceptually) as a group of concentric circles
- Bcan be read or written with a single input/output operation by the controller
135A sector usually has at least two parts: a preamble and a data region. Which of the following is usually found in the preamble?
- Ainformation identifying the physical location of the sector on the disk
- Bthe user ID of the user whose process actually wrote the data.
136The rotational speed of a disk (when it is operating) is usually
- Aconstant, with some tolerance for variation
- Bcompletely variable, with lower speeds being selected to reduce power consumption onmobile devices.
137A particular disk rotates at 7200 revolutions per minute (which is a common speed). Once the read/write heads have reached the proper cylinder, what is theaveragerotational delaythat will be encountered before data transfer can begin? Assume each sector on the disk is aslikely to be read as any other
- A1/7200 second
- B1/240 second
138What is the path taken by the electron beam in a CRT used in a video monitor?
- AThe path is top to bottom on the leftmost column, then top to bottom on the next column to the right, and so forth, and finally back to the upper left corner where the cycle repeats
- BThe path is left to right on each row, then on to the beginning of the next lower row, ...,and finally back to the upper left corner where the cycle repeats.
139In early systems, the image to be displayed on a video display was
- Astored in a mercury delay line.
- Bstored in the computer's primary memory
140In modern systems, the image to be displayed on a video display is located
- Ain a mercury delay line
- Bin special memory that is located in the video adapter
141The acronym PIXEL standard for
- Apicture element.
- Bpicture electrical line
142What does it mean for memory to be dual ported?
- AEach read of a byte in the memory is also accompanied by a write operation.
- BThe memory can be accessed (read and written) by the CPU and a device controller at the same time without introducing excessive delay.
143The acronym DMA stands for
- ADirect Memory Access
- BDual Memory Access
144When DMA is used
- Athe CPU, after setting up the DMA, is not directly involved in the transfer of every data byte between memory and an input/output device
- Bdevice drivers are simpler
145The purpose of interleaving (as applied to disks) is
- Athe logical reordering of the sectors to reduce rotational latency
- Bthe multiplexing of data from two physical sectors on a disk to the disk controller.
146Input/output software is designed in layers. Each layer
- Aprovides services to the layer above it, and uses services provided by the layer below it.
- Bdeals with only a single device (or device type).
147Early operating systems (like CP/M)
- Ahad only a single system call for input/output, with parameters identifying the device, the operation to be performed, the memory location of the data, and the number of bytes to be processed.
- Bhad separate system calls for input/output on serial ports, keyboard, displays, and disk drives
148What should normally be done if a read operation on a disk produces a physical error?
- AThe device should be marked "off line" and the operator should be notified; the process requesting the read should remain blocked.
- BThe operation should be repeated, perhaps a few times, because the error might be transient
149The goal of concurrency in input/output systems is
- Ato allow a process to continue operation while one or more input/output operations are in progress.
- Bto allow the CPU to be accessing several different physical devices at the same time
150Which of the following devices are usually not shared by processes from several users?
- Aa serial port
- Ba disk drive
151When a disk drive completes a read or write operation, it is common for
- Aan interrupt to be generated
- Bthe process requesting the operation to be blocked
152Which of the following events is likely to be able to generate an interrupt?
- Aa byte has been written to a disk drive.
- Ba byte received on a serial port has bad parity
153In a system that uses memory-mapped input/output, registers on device controllers appear as
- Aordinary memory locations that may be read or written.
- Bsemaphores.
154The first task of the interrupt handler for a device is
- Ato obtain information from device controller registers to determine the reason for the interrupt.
- Bto start another input/output operation.
155A device driver rarely completes all the work associated with an interrupt from a device's controller. Instead, it will usually
- Aawaken a blocked process to complete the work
- Bcreate a new process to handle the rest of the work
156When several devices share the same interrupt, the device driver that handles the interrupt
- Amust poll the devices to determine which of them caused the interrupt.
- Bstops input/output on all but one of the devices, and restarts the operation on the remaining device
157The initialization routine (or component) of a device driver
- Ais invoked during system startup to determine if the device exists, and if so, how many units of the device are present
- Bduring system shutdown to place the device into a consistent state
- Cz
158The strategy routine in a device driver
- Adetermines the best data path through a DMA channel
- Bqueues requests for a block device until they can be processed.
159A watchdog timer
- Ais not commonly used by device drivers in modern operating systems.
- Bsignals a device driver if an input/output operation does not complete in a reasonable time.
160The term logical block refers to
- Athe reason a process will block while waiting on an interrupt from an input/output device's controller
- Ba group of consecutive physical blocks, or sectors, treated as a single unit for efficiency
161The system buffer pool or buffer cache is used to
- Ahold the identities of all processes that are blocked waiting on a disk input/outputoperation to complete
- Bhold copies of blocks that have been read from disk.
162It is common for application languages (like C, C++, and Java) to provide libraries of input/output functions that perform the actual system calls for input/output. When these library functions read or write a disk, for example, they
- Arequest data transfers in blocks, rather than in single bytes.
- Brequest transfers directly to or from the disk, rather than using the system buffer cache
163If there is only a single process executing in a system and no concurrent input/output is permitted, the disk device drive will process requests
- Ausing the FIFO algorithm.
- Busing the LOOK algorithm
164The SSTF algorithm for disk scheduling orders requests
- Ato minimize seek time
- Bto minimize transfer time
165The SCAN algorithm is one of a class of disk scheduling algorithms characterized by the name
- Aelevator algorithms
- Bfinite state algorithms.
166Assume the SCAN algorithm is in use, the heads are moving up, are currently processing a request on cylinder 20, and the upsweep queue contains 25, 37, and 58. A new requestarrives for cylinder 62. It will be placed
- Aafter 58 on the upsweep queue
- Bat the tail of the downsweep queue
167Assume the SCAN algorithm is in use, the heads are moving up, are currently processing arequest on cylinder 20, and the upsweep queue contains requests for cylinders 25, 37, and58. A new request arrives for cylinder 22. It should be placed
- Aon the downsweep queue
- Bbefore cylinder 25 on the upsweep queue
168Assume the SCAN algorithm is in use, the heads are moving up, are currently processing a request on cylinder 20, and the upsweep queue contains 25, 37, and 58. After processing requests on which cylinder will the algorithm begin a downsweep?
- AIt is impossible to tell with the information given.
- BNone of the other answer choices is correct
169The primary difference between the SCAN and the LOOK disk scheduling algorithms is
- ASCAN moves between the lowest and highest physical cylinder numbers, and LOOK moves between the lowest and highest cylinder numbers at which requests are located
- BLOOK defers new requests for the cylinder on which the heads are located, but SCAN adds them to the current queue (that is, the upsweep or downsweep queue that is being processed).
170The primary difference between the SCAN and C-SCAN disk scheduling algorithms is
- AThe SCAN algorithm numbers cylinders starting with 1, but the C-SCAN algorithm numbers them starting with 0
- BSCAN maintains an upsweep and a downsweep queue, but C-SCAN treats the disk as if it were a cylinder, with the lowest numbered cylinder being processed immediately after the highest numbered cylinder
171Which of the following is the correct meaning of the RAID acronym?
- ARedundant Array of Inexpensive Disks
- BRedundant Array of Information Disks
172Modern disk controllers
- Aanticipate the direction in which the heads will move next, and start moving them in that direction even before the next command is given
- Bhave sufficient memory to store the contents of an entire disk track
173Most modern computer systems maintain a battery-powered clock that can provide the date and time. When is this clock normally read?
- Aat least once every hour
- Bonly when the system is booted
174A programmable timer usually has two registers. One register is decremented each time a fixed frequency oscillator goes through one cycle. When this register reaches zero
- Aan interrupt is generated
- Bthe currently running process is terminated.
175A programmable timer usually has two registers. One register is decremented each time a fixed frequency oscillator goes through one cycle. The other register
- Ais incremented each time the first register reaches zero
- Bis often used to reload the first register when it reaches zero
176Since a system has a limited number of hardware timers, and many timers may be needed by processes, most systems maintain a queue of virtual timers. These virtual timers are ordered by
- Athe time at which the virtual timers will reach zero
- Bthe inverse order in which the virtual timers were created
177Suppose the virtual timer queue in a system has timers scheduled to expire in 10, 20, and 30 seconds from the current time. The entries on the virtual timer queue will have delta times of
- A30, 20, and 10
- B10, 10, and 10
178Suppose the delta times on the virtual timer queue, in order, are 5, 10, and 5. A new virtual timer is required for time 18. What will the delta times be, in order, after the new request is added to the queue?
- A5, 10, 3, 2
- B5, 10, 5, 18
179A terminal is the term used for several separate devices that are grouped together. What are those devices?
- Aa keyboard and a display
- Ba keyboard and a mouse
180An RS-232 terminal is so named because
- Ait uses a standard (RS-232) serial connection between the terminal and a computer system (or modem)
- Bit can receive and send (RS) 232 characters per second
181An X terminal is so named because
- Ait includes an X-mouse
- Bit uses the X protocol for communication
182An escape sequence is
- Aa sequence of characters used to bypass the password prompt issued by a computer system
- Ba string of characters, usually beginning with an escape character, that requests a special command for a terminal.
183The upper limit on serial data rates (as used with terminals) is about
- A56,000 bits per second.
- B8192 bits per second
184When a terminal operates incookedmode, each character typed by the user
- Ais displayed
- Bis checked to see if it is part of an escape sequence.
185When a terminal operates inrawmode, each character typed by the user
- Ais checked to see if it is a control-C
- Bis passed immediately to the application without examination by the operating system
186 One character code sent to a terminal that may require processing by a non-terminal device is
- Aan ALT character.
- Ba bell character
187ASCII is a standard for codes associated with the characters commonly found on terminals. A major problem with ASCII is
- Ait omits many characters used internationally
- Bit has no support for control characters
188Why might a system treat memory allocation requests from the operating system itself differently from those generated by an application?
- AThe OS kernel is running in protected mode, and the application is running in user mode
- BFailure of a memory request made by the OS could possibly cause the entire system to fail. Failure of an application request would cause a single application to fail, not the entire system.
189Since memory is a non-consumable resource
- Athere must be facilities to deallocate (reclaim) previously used regions of memory when they are no longer used.
- Bprocesses must never request more memory than is currently available.
190The sole advantage of monoprogramming over multiprogramming is
- Ait eliminates the need for mutual exclusion, thereby eliminating the possibility of deadlock
- Bsimplicity
191What is done with memory that is not used by the single process executing in a monoprogrammed system?
- AIt is used to provide interprocess communication
- BIt is unused, and therefore wasted
192Which of the following is required of a program designed to run on a monoprogrammed system?
- AIt must be written so it can use a variable amount of memory
- BIt cannot use more memory than is physically available.
193Which of the following operating systems was the predecessor of Microsoft MS-DOS?
- ACP/M
- BDynix
194Which of the following is the name of a technique that makes it possible to have multipleprocesses resident in memory at the same time?
- Amultiprogramming
- Bmultiloading
195Which of the following techniques can be used to perform a reasonably accurate analysis of the behavior of multiprogramming systems?
- Abirth and death Minkowski model
- Bbirth and death Markov process model
196In modeling multiprogramming systems, each state in the model represents a unique number of processes in the ready state, and state 0 corresponds to the system being in a state with all processes blocked (not ready). How many ready processes are represented by state 1 in the model?
- Aone or more processes
- Bexactly one process
197The simple model used to determine processor utilization discussed uses elementary probability theory to determine the probability that all N processes are not blocked when the probability of an individual process being blocked is p. What is the probability that all N processes are not blocked?
- A(p to the Nth power) - 1
- B1 minus (p to the Nth power)
198In general, what happens to processor utilization as the number of processes in memory increases but the probability of a process being blocked remains constant?
- AProcessor utilization increases to a point, then begins to decrease.
- BProcessor utilization increases.
199Assume the number of processes resident in memory remains constant, but the probability of each process being blocked increases. In general, what happens to overall processor utilization?
- AProcessor utilization decreases.
- BProcessor utilization increases.
200In multiprogramming we place multiple processes in memory at the same time. Where these processes are placed varies. In the IBM operating system named OS/MFT, where were processes placed?
- AEach process was placed in one of several regions of memory, each with a fixed size.
- BEach process is placed immediately after the region of memory associated with the lastprocess placed in memory
Information
- Description
- Quiz on Quiz 2, created by Myles Allbee on 04/05/2016.
- Author
- Myles Allbee
- Language
- English
- Views
- 2
- Rating
- No ratings yet Rate this quiz