Deck 17: Understanding Distributed Systems and Operating Systems

Question 1

Which of the following statements is false?

A) Distributed systems improve the capacity and reliability of a single machine.
B) Users of a distributed operating system are given the impression that they are interacting with just one machine.
C) Distributed systems divide resources among a group of machines and let multiple users share the resources.
D) Distributed systems are easy to implement.

Answer 1

Distributed systems are easy to implement.

Question 2

__________ allows a distributed system to grow without affecting the existing applications and users.

A) Transparency
B) Scalability
C) Reliability
D) all of the above

Answer 2

Scalability

Question 3

Operating systems provide persistence transparency by __________.

A) hiding the location of resources
B) hiding the fact that multiple copies of a resource are available
C) hiding the location where the resource is stored
D) hiding the movement of components among computers in a distributed system

Answer 3

hiding the location where the resource is stored

Question 4

Network operating systems__________.

A) access remote resources transparently
B) are easier to implement than distributed operating systems
C) manage resources located in multiple networked computers
D) none of the above

Question 5

__________ is not one of the eight types of transparency a distributed system can provide.

A) access tranparency
B) failure tranparency
C) security tranparency
D) transaction tranparency

Question 6

Middleware helps provide __________.

A) portability
B) transparency
C) interoperability
D) all of the above

Question 7

In RPC, packaging procedure arguments along with the procedure name into a message for transmission over a network is performed by __________.

A) packaging
B) marshaling
C) the client stub only
D) the server stub only

Question 8

The __________ is not one of the three layers of RMI architecture.

A) presentation layer
B) stub/skeleton layer
C) remote reference layer
D) transport layer

Question 9

_________ is an open standard designed to enable interoperation among programs in heterogeneous as well as homogeneous systems.

A) RPC
B) RMI
C) CORBA
D) DCOM

Question 10

_________ moves a copy of a process to another machine without destroying the original process.

A) Process migration
B) Process cloning
C) Process distribution
D) none of the above

Question 11

In distributed systems_________ make(s) it difficult to determine the order in which events occur.

A) unpredictable communication delays
B) temporal transparency
C) process migration
D) the absence of middleware

Question 12

A __________ ensures that all events are ordered and that causality is preserved.

A) causal ordering
B) partial ordering
C) total ordering
D) logical ordering

Question 13

__________ guarantees that when multiple messages are sent between processes, the messages will arrive in the order they were sent.

A) Atomic broadcast
B) FIFO broadcast
C) Totally ordered broadcast
D) Causal broadcast

Question 14

Agrawala and Ricart's distributed mutual exclusion algorithm __________.

A) assumes that communication between processes is reliable
B) assumes that processes do not fail
C) requires that a process first send a request message to all other processes in the system and receive a response from each of them before that process can enter its critical section
D) all of the above

Question 15

A phantom deadlock __________.

A) is an example of resource deadlock
B) occurs in the presence of circular waits for communication signals
C) occurs when a process that holds a shared resource fails
D) none of the above.

Question 16

Which of the following statements about wound-wait and wait-die deadlock prevention strategies is true?

A) The wound-wait deadlock prevention strategy prevents deadlock by denying the wait-for condition.
B) The wait-die deadlock prevention strategy breaks deadlock by denying the no-preemption condition.
C) In the wound-wait strategy, as a process ages, it will wait less often.
D) In the wait-die strategy, as a process ages, it will wait less often.

Question 17

Order the distributed deadlock detection strategies with the least fault tolerant first.

A) central, hierarchical, distributed
B) hierarchical, central, distributed
C) central, distributed, hierarchy
D) distributed, hierarchical, central

Question 18

In a transaction wait-for graph (TWFG), each node represents a __________ and each arrow represents a __________.

A) process, resource
B) resource, resource allocation
C) process, requested resource being held by another process
D) resource, deadlock

Question 19

Which of the following statements about the Sprite distributed operation system is false?

A) Sprite uses process migration to balance the workload of the system.
B) Sprite uses explicit acknowledgements to ensure that messages are delivered reliably.
C) Location-independent calls produce the same result for all workstations.
D) The Sprite file system caches files on both the server and client.

Question 20

Which of the following statements about the Amoeba distributed operation system is true?

A) Each Amoeba user is assigned to a particular workstation for the duration of the session.
B) Amoeba does not provide location transparency.
C) Amoeba uses implicit acknowledgement to ensure that messages are delivered reliably.
D) Files stored by the bullet server cannot be modified after being created.