UNIVERSITY OF MUMBAI

MCQ QUESTION BANK (100 Questions)

Course Code and Name:CSC802 (Distributed Computing)
Class:BE Department :COMPUTER
Semester:VIII

OPTIONS
Module No Weighatge in Hrs Q NO QUESTION ( 2 marks per question) A B C D Correct Answer
1 6 A distributed system is defined as a
collection of autonomous computers
1 linked by a network with software TRUE FALSE TRUE
designed to produce an integrated
computing facility.
A “glue” between client and server parts of
2 Middleware Firmware Package System Software Middleware
application
uniprocessor
multiprocessor multiprocessor uniprocessor
3 The hardware of DS has two types system,multicomputer multiprocessor system,multicomputer system
system,multicomputer system system,unicomputer system system,unicomputer system
system
loss of messages occurs loss of messages occurs
loss of messages occurs between loss of messages occurs
between the incoming between the incoming loss of messages occurs between the incoming
4 Channel omission failure is the sending process and the between the incoming buffer
message buffer and the buffer and the outgoing buffer and the outgoing buffer
outgoing message buffer. and the system.
receiving process. buffer
Can be detected in
Process Fail - Stop in process omission Can be detected in synchronous Can be detected in Can be detected in
5 synchronous and Can be detected in synchronous system.
faults system. asynchronous system. standalone system.
asynchronous system.
Distributed pervasive system is also
6 Ubiquitous computing User Interface design Graphical User interface peer to peer system Ubiquitous computing
known as
Transaction Processing System Transaction Product System Transparency Processing Transaction Processing
Two forms of distributed information Transaction Processing System and Enterprise
7 and Enterprise Application and Enterprise Application System and Independent Software and Enterprise
systems are: Application Integration.
Integration. Interrogation. Application Integration. Access Integration.
8 Type of cluster computing is Load sharing cluster Load holding cluster Load replication cluster Load balancing cluster Load balancing cluster
9 Type of grid computing is Collaborative Grid System Grid Process Grid Channel grid Collaborative Grid
10 Scaling transparency hides System expansion System collaboration System failure System security System expansion

OPTIONS
Module No Weighatge in Hrs Q NO QUESTION ( 2 marks per question) A B C D Correct Answer
2 10 1 Internet provides _______ for remote
telnet http ftp RPC telnet
login.
2
The header usually consists of the
following elements in IPC I. Sequence
I,III, IV I,II, III I, II, IV I,IV I,II, III
number II. Structural information III.
Address IV. File mode(Read/ Write)
3 In RMI, the objects are passed by
Value Reference Value and Reference object Value
_____.
4 What are the exceptions which have
Class-Not-Found-Excepti
to be handled in a RMI client MalFormedURLException NotFoundException ArithmeticException MalFormedURLException
on
program?
5 Transient communication done
Network Transport Physical Session Transport
by_____ layer
6 The local operating system on the
server machine passes the incoming server stub client stub client operating system Binding agent server stub
packets to the _________
7 Machine that places the request to
access the data is generally called as Server Machine Client Machine Request Machine Database server Client Machine
__________.
 8 ________________ provides
programmers a familiar programming
Permanent procedure
model by extending the local Distributed environment Process and file Remote procedure call Remote procedure call
call
procedure call to a distributed
environment
9 An architecture where clients first
communicate the server for data then
Client/Server architecture Three-tier architecture Two-tier architecture Peer-to-Peer architecture Client/Server architecture
format and display it to the users, is
known as ___________
10 _________ is an object acting as a
skeleton stub remote server stub
gateway for the client side.
11 Message passing provides a
by sharing the same
mechanism to allow processes to By sharing the same address without sharing the same
process number and By sharing port number By sharing the same address space
communicate and to synchronize their space address space
Process Identifier
actions ________
12 Which of the following
RRL Stub/skeleton layer Transport layer Networking layer RRL
allocates/deallocates buffers
13 optical service
OSI stands for __________ open system interconnection operating system interface open service Internet open system interconnection
implementation
14 Which address is used to identify a process
physical address logical address port address specific address port address
on a host by the transport layer?
15 response from
In all reliable multicast group n' response expected from the response from any of the response needed from all
'm'(1<m<n) of the 'n' response needed from all the receivers
communication ______ receiver receiver required the receivers
receiver required
16 If processes p and q both receive
messages m and m', then p receives
m before m' if and only if q receives m Absolute ordering Consistent ordering Causal ordering FIFO ordering Consistent ordering
before m'.The order delivery is
called______________
17 processes that are not
Only members can send messages processes that are not
What is close group in group the idea of groups is to members (clients) but close Only members can send messages to the
to the group as a whole members (clients) can send
communication? support replicated servers to the group can send group as a whole
message to the group.
message to the group.
18 Message queuing systems is design for tightly coupled loosely coupled
transient communication synchronous communication loosely coupled communication
__________ communication communication
19 connection-less connection-oriented
Berkeley socket is used for ____ group communication Causal communication connection-oriented communication
communication communication
20 It is responsible for It is responsible for It is responsible for
what is the task of RPCRuntime ? It is responsible for marshaling It is responsible for retransmission,
retransmission handling exception syntactic transparency
21
_________ all related objects moved
Call by value Call by move Call by visit Call by reference Call by move
and left to a server upon the first RPC
22 Longer server recovery time File operations must be Longer server recovery time
What is the feature of stateful server? Quick recovery after reboot Simple server design
idempotent

QUESTION ( 2 marks per question) OPTIONS Correct Answer

Module No Weighatge in Hrs Q NO A B C D
3 10 In which algorithm, One process is elected Distributed mutual exclusion Centralized mutual
1 Token ring algorithm Lamport algorithm Centralized mutual exclusion algorithm
as the coordinator. algorithm exclusion algorithm
_____ is a process that prevents multiple
2 threads or processes from accessing shared Critical section Deadlock Message passing Mutual Exclusion Mutual Exclusion
resources at the same time.
Absolute time synchronization can be
3 Vector time stamping method Christian’s method Lamport’s method Ricart-Agrawala algorithm Christian’s method
achieved using
More efficient and more fault More efficient but less fault Less efficient but more fault Less efficient and less fault
The Ricart & Agrawala distributed mutual Less efficient and less fault tolerant than a
4 tolerant than a centralized tolerant than a centralized tolerant than a centralized tolerant than a centralized
exclusion algorithm is: centralized algorithm
algorithm. algorithm algorithm algorithm
 Which mutual exclusion algorithm is
5 useful when the membership of the group Centralized Lamport’s. Token ring Decentralized Algorithm Centralized
is unknown?
More efficient and more fault Less efficient and less fault Less efficient but more fault More efficient but less fault
The Ricart & Agrawala distributed mutual Less efficient and less fault tolerant than a
6 tolerant than a centralized tolerant than a centralized tolerant than a centralized tolerant than a centralized
exclusion algorithm is centralized algorithm
algorithm. algorithm algorithm. algorithm.
Which event is concurrent with the vector
7 (3,9,5) (3,8,4) (1,7,3) (4,8,2) (4,8,2)
clock (2, 8, 4)?
A client gets a timestamp of 4:12:30.500
from a
time server. The elapsed time between the
request
and response was 20 msec (0.020 sec).
8 4:12:30.480 4:12:30.490 4:12:31.510 4:12:31.520 4:12:31.510
The current
time on the client is 4:12:30.510. Using
Cristian’s
algorithm, what is the time set to on the
client?
9 NTP is __________ layer protocol. Application session transport physical Application
Which of the following is an example of
10
election algorithm. Berkley Algorithm: Bully Algorithm. Cristian’s Algorithm LAMPORT’S Bully Algorithm.
Having all nodes in a ring of Sending a message around Building a list of all live
processors send a message to a Sending a token around a set all available nodes and nodes and choosing the
11
coordinator who will elect the of nodes. Whoever has the choosing the first one on largest numbered node in the Building a list of all live nodes and choosing
The ring election algorithm works by: leader. token is the coordinator. the resultant list. list. the largest numbered node in the list.
For each critical section (CS) execution,
Ricart-Agrawala algorithm requires
12 ___________ messages per CS execution
and the Synchronization delay in the
algorithm is ____________. 3(N − 1), T 2(N − 1), T (N − 1), 2T (N − 1), T 2(N − 1), T
For each critical section (CS) execution,
maekawa’s algorithm requires
13 ___________ messages per CS execution
and the Synchronization delay in the
algorithm is ____________. √N , T 2√N , T 3√N , T 3√N , 2T 3√N , 2T
Bully Algorithm was proposed by
14
____________ Garcia-Molina.
RAYMOND'S TREE BASED Centralized Based physical clock
15
ALGORITHM is an ___ Non- token based algorithm. Token based algorithm. algorithm synchronization algorithm. Token based algorithm.
Suzuki-Kasami's Broadcast Algorithm is Centralized Based physical clock
16
an Non- token based algorithm. Token based algorithm. algorithm synchronization algorithm. Token based algorithm.
Network Timestamp
17
Full form of NTP is: Network Time Protocol New time Protocol New Timestamp Protocol Protocol Network Time Protocol
Which algorithm requires "1 to ∞"
18 messages to enter and leave a critical
region? Token ring algorithm Centralized algorithm Decentralized algorithm Distributed algorithm Token ring algorithm

OPTIONS
Module No Weighatge in Hrs Q NO QUESTION ( 2 marks per question) A B C D Correct Answer
4 10 Static load balancing Symmetric load balancing Asymmetric load balancing
1
Dynamic load balancing algorithm algorithm
In _______ assignment of job is done at runtime algorithm algorithm Dynamic load balancing algorithm
2 Pretransfering also known as__ Premigrating Precopying Prefiltering Postcopying Precopying
3 run()to runnable
____ change the state of thread from suspended yield() destroy() start() start()
4 Printer
Distributed syatem consists of set of resources Processor
interconnected by network. Resources means___ CD Processes Processor
5 This is not feature of cooperative algorithm complex larger overhead worst stability better stability worst stability
How is access to resources of various Zone are configured for
6
machines is done? Remote logging using ssh or telnet automatic access FTP is not used FTP is used Remote logging using ssh or telnet
 7 What are the characteristics of data transfer data by entire file or transfer the computation execute an entire process or execute an entire process or transfer data by entire file or immediate
migration? immediate portion required rather than the data parts of it at different sites parts of it at same site portion required

8 What are the characteristics of transfer data by entire file or transfer the computation execute an entire process or execute an entire process or
computation migration? immediate portion required rather than the data parts of it at different sites parts of it at same site transfer the computation rather than the data

9 transfer data by entire file or transfer the computation execute an entire process or execute an entire process or execute an entire process or parts of it at
immediate portion required
What are the characteristics of process migration? rather than the data parts of it at different sites parts of it at same site different sites
When the process issues an I/O request It is placed in a waiting It is placed in the ready
10
__________ It is placed in an I/O queue queue queue It is placed in the Job queue It is placed in an I/O queue

OPTIONS
Module No Weighatge in Hrs Q NO QUESTION ( 2 marks per question) A B C D Correct Answer
5 8 Absolute time ordering of all shared
1 Sequential Consistency Casual Consistency Strict Consistency Weak Consistency Strict Consistency
accesses matters in
In Casual consistency model all processes
2 random order same order sequential order specific order same order
shared accesses in
In which of the following consistency
3 model all writes become perceptible to all Strict Weak Casual Sequential Strict
processes
________ consistency is that write
operations by the same process are
4 Weak Strict Eventual FIFO FIFO
performed in the correct order
everywhere.
Any successive write operation by a
process on a data item x will be performed
5 Monotonic-write Writes follows reads Read your writes Monotonic-read Writes follows reads
on a copy of x that is up to date with the
value most recently read by that process.
__________replicas are used only to
6 Client initiated Server initiated Permanent Dynamic Client initiated
improve access time on data
In ________________receivers never
acknowledge the successful delivery of Non-hierarchical Hierarchical feedback
7 Basic reliable multicasting Atomic multicast Non-hierarchical feedback control
multicast message but instead report only feedback control control
when missing the message.
A problem with the ________ protocol is
that when the coordinator has crashed,
8 One-phase commit Three-phase commit Two-phase commit Virtual synchrony Two-phase commit
participants may not be able to reach a
final decision.
Optimistic logging protocols need to keep
9 Independencies Checking points Dependencies Erroneous state Dependencies
track of
All write accomplished
All write accomplished on All write accomplished on All write accomplished on
Processor Consistency model on random memory All write accomplished on identical
10 identical memory location in random memory location identical memory location
guarantees and conforms that location in identical memory location in identical sequence.
identical sequence. in random sequence in random order
sequence
Remote write protocol supports all Need to be forwarded to local Need to be forwarded to Need to be forwarded to Need to be forwarded to Need to be forwarded to a fixed single
11
write operations server multiple servers. any server a fixed single server. server.
Staleness deviations relate to the
12 First time Most recent time last time previous time last time
_________ a replica was updated.
more of management
13 The placement of replica servers is optimization problem consistency performance more of management issue
issue
The dynamic replication algorithm files on server can be schedule process
14 to reduce load on server resource sharing to reduce load on server
takes into account migrated anywhere migration
to improve resource to improve access time to improve the
15 Client Initiated Replicas are used to improve load balancing to improve access time to data
sharing to data performance
Server reacts
16 State transition failures happens Server fails Client fails Network fails Server reacts unexpectedly
unexpectedly
 The main problem with the reliable Support large number of Support fixed number of cannot support fixed cannot support large cannot support large numbers of
17
multicast scheme is that receivers receiver numbers of receivers. numbers of receivers. receivers.
In Erasure correction approach a other successfully delivered
18 last delivered packet first packet retransmitting packets other successfully delivered packets
missing packet is constructed from packets
combine distributed
To improve performance, many combine check pointing with combine check pointing combine check pointing combine check pointing with message
19 commit with message
distributed systems recovery oriented computing with message logging with distributed commit logging
logging.
coupled in the sense that decoupled in the sense decoupled in the sense
Rebooting as a practical recovery coupled in the sense that there decoupled in the sense that there are few
there are no that there are few or no that there are
20 technique requires that components are dependencies between or no dependencies between different
dependencies between dependencies between dependencies between
are largely different components. components
different components. different components different components.
OPTIONS
Module No Weighatge in Hrs Q NO QUESTION ( 2 marks per question) A B C D Correct Answer
6 12 1 File Replication is done to _____ increase complexity increase cost increase reliability increase data increase reliability
in NFS which funtion is used for creating
open() create() develope() null() create()
2 new file?
which command is used to create a
rmdir symlink mkdir open mkdir
3 directory
datanodes and namenode are two elements
AFS HDFS NFS None of the above HDFS
4 of which file system?
In which file system mapreduce function is
AFS NFS HDFS None of the above HDFS
5 used?
In distributed file system, DNS stands for? Domain Name System Domain Name Server Directory name service Disk name system Domain Name System
6
7 NFS file System uses ______Mechanism RPC CORBA RMI None of the above RPC
8 map and reduce are____ libraries functions file system OS functions
In HDFS file System, A ________ serves
as the master and there is only one Data Node NameNode Replication Data block NameNode
9 NameNode per cluster
Data blocks are replicated DataNode is the Data blocks are replicated
For HDFS file system, Point out the Each incoming file is broken into across different nodes in the slave/worker node and across different nodes in the DataNode is the slave/worker node and holds
correct statement. 32 MB by default cluster to ensure a low holds the user data in the cluster to ensure a low the user data in the form of Data Blocks
10 degree of fault tolerance form of Data Blocks degree of fault tolerance
11 HDFS works in a __________ fashion. master-master master-slave slave-slave None of the above master-slave
In HDFS file System, ________
NameNode is used when the Primary Rack Data Secondary primary Secondary
12 NameNode goes down.
Replication Factor can be Block Report from each
User data is stored on the DataNode is aware of the
In context of HDFS file system, Point out configured at a cluster level DataNode contains a list of DataNode is aware of the files to which the
local file system of files to which the blocks
the wrong statement. (Default is set to 3) and also at a all the blocks that are stored blocks stored on it belong to
DataNodes stored on it belong to
13 file level on that DataNode
In HDFS, . ________ is the slave/worker
node and holds the user data in the form of DataNode NameNode Data block Replication DataNode
14 Data Blocks.
15 In distributed file system,XDR stands for? external data request External device request external data recovery External data representation External data representation
If file system is growing without affecting
performance of the system then this Union portable robust Scalability Scalability
16 feature is called as?
Storing file in________ makes it
Secondary Memory RAM Register DRAM Secondary Memory
17 permanently available
Session file sharing semantics are suitable
complete folder complete file One byte one block complete file
18 for caching
Network file system(NFS) is developed
Sun Microsystem oracle apple honeywell Sun Microsystem
19 by?
The file once created can not be changed is
Rigid file REX file Immutable file Robust file
20 called ___________ C