41 + Mcqs in Multiprocessors Page 1 McqOptions

1.	The directory must be tracking the group of nodes which have a copy of the block; then the used set is called
A.	Dedicators
B.	Sharers
C.	Host
D.	Distributors
Answer» C. Host

Discussion

2.	The particular block's statuses of physical memory are normally kept in one location, called
A.	Register
B.	Directory
C.	Stack
D.	Queue
Answer» C. Stack

Discussion

3.	The pair of instructions including a special load known as a load-linked or load-locked with a special store called a
A.	Store linked
B.	Store locked
C.	Store conditional
D.	Store operational
Answer» D. Store operational

Discussion

4.	In the coherent multiprocessor, caches that is present provides both the migration and
A.	Coherence
B.	Recurrence
C.	Replication
D.	Uniformity
Answer» D. Uniformity

Discussion

5.	Cases where variables gets updated without the order of the synchronization are called
A.	Dedicators
B.	Sharers
C.	Data races
D.	Share races
Answer» D. Share races

Discussion

6.	One assigned operation for building synchronized operations, is called the
A.	Atom exchange
B.	Atomic exchange
C.	Remote node
D.	Both a and b
Answer» C. Remote node

Discussion

7.	All writing procedures for same location are seen having same order; the stated property is called
A.	Write serialization
B.	Read serialization
C.	Parallel processing
D.	Synchronization
Answer» B. Read serialization

Discussion

8.	For achieving a speedup up to 80 with 100 processors. What original computation fraction can be sequential
A.	0.875
B.	0.9975
C.	1.9975
D.	2.9975
Answer» C. 1.9975

Discussion

9.	One that programming model which allows for having a more efficient implementation, is to suppose that programs are
A.	Remote
B.	Synchronized
C.	Atomic
D.	Shared
Answer» C. Atomic

Discussion

10.	Multiple-applications independently running, are typically called
A.	Multithreading
B.	Multiprogramming
C.	Multitasking
D.	Synchronization
Answer» C. Multitasking

Discussion

11.	The requesting node sending the requested data starting from the memory, and the requestor which has made the only sharing node, known as
A.	Write miss
B.	Write node
C.	Read miss
D.	Read node
Answer» D. Read node

Discussion

12.	Server clusters grow to 10th of thousands and beyond, are called
A.	Warehouse-scale computer
B.	Servers
C.	Kernel
D.	Supercomputers
Answer» B. Servers

Discussion

13.	The tightly coupled set of threads' execution working on a single task, that is called
A.	Multithreading
B.	Recurrence
C.	Parallel processing
D.	Serial processing
Answer» D. Serial processing

Discussion

14.	Microprocessors which are directly connected memory to a single-chip, that is sometimes called as
A.	Backside
B.	Memory bus
C.	Kernel
D.	both a and b
Answer» E.

Discussion

15.	From inter-processor communication, the misses arises are often called
A.	Coherence misses
B.	Commit misses
C.	Parallel processing
D.	Hit rate
Answer» B. Commit misses

Discussion

16.	When every cache hierarchy level is a subset of level which is further away from the processor, is refered to as
A.	Multilevel inclusion
B.	Synchronization
C.	Atomic synchronization
D.	Distributors
Answer» B. Synchronization

Discussion

17.	The node which has the memory location and the entry of directory of an address is
A.	Home node
B.	Guest node
C.	Host node
D.	All above
Answer» B. Guest node

Discussion

18.	Relaxing the W?R ordering, will yield a model known as
A.	Total store ordering
B.	Processor consistency
C.	Link locks
D.	Both a and b
Answer» E.

Discussion

19.	An operation being done in a way that intervening operation can be occurred, this operation is called a
A.	Atomic
B.	Molecular
C.	Multitasking
D.	Serial processing
Answer» B. Molecular

Discussion

20.	Fetch-and-increment is another task of the
A.	Atom exchange
B.	Atomic exchange
C.	Atomic synchronization
D.	None of above
Answer» D. None of above

Discussion

21.	The protocols for maintaining coherence of multiple processors are known as
A.	Data coherence protocols
B.	Commit coherence protocols
C.	Recurrence
D.	Cache coherence protocols
Answer» E.

Discussion

22.	Private data that is used by a single-processor, then shared data are used by
A.	Single processor
B.	Multi processor
C.	Single tasking
D.	Multi tasking
Answer» C. Single tasking

Discussion

23.	Relaxing the W?W ordering, will yield a model known as
A.	Total store ordering
B.	Partial store order
C.	Link locks
D.	Processor consistency
Answer» C. Link locks

Discussion

24.	To update the cached copies of the data item; is the alternative protocol which is known as
A.	Write update
B.	Write broadcast protocol
C.	Read protocol
D.	both a and b
Answer» E.

Discussion

25.	The term which is named as shared memory with both SMP and DSM referring that the address space is
A.	Multi-threading
B.	Recurrence
C.	Dedicated
D.	Shared memory
Answer» E.

Discussion

26.	Symmetric multiprocessors architectures, are sometimes known as
A.	Uniform memory access
B.	Static memory access
C.	Variable memory access
D.	All above
Answer» B. Static memory access

Discussion

27.	Two-way set associative having a 64-byte block, the single clock-cycle hit time is a
A.	Level 1 instruction cache
B.	Level 1 data cache
C.	Level 2 data cache
D.	Level 2 instruction cache
Answer» B. Level 1 data cache

Discussion

28.	When the home node being the local node, the copies may exist in a third node, called
A.	Home node
B.	Guest node
C.	Remote node
D.	Host node
Answer» D. Host node

Discussion

29.	A remote node is being the node which has a copy of a
A.	Home block
B.	Guest block
C.	Remote block
D.	Cache block
Answer» E.

Discussion

30.	The alternative design technique consists of multiprocessors having physically distributed memory, called
A.	Warehouse-scale
B.	Static memory access
C.	Distributed shared memory
D.	Shared memory
Answer» D. Shared memory

Discussion

31.	Only one node having a cache block copy, and this cache has written the block, and this memory copy is out of date. Then the processor is called the
A.	Host
B.	Owner
C.	Guest
D.	Server
Answer» C. Guest

Discussion

32.	Plotting performance versus number of processors, is refered to as
A.	Mortar shot graphs
B.	Flow chart
C.	K-Map
D.	Plotters
Answer» B. Flow chart

Discussion

33.	The alternative way of a snooping-based coherence protocol, is called a
A.	Memory protocol
B.	Directory protocol
C.	Register protocol
D.	None of above
Answer» C. Register protocol

Discussion

34.	The main objective in building the multimicroprocessor is
A.	greater throughput
B.	enhanced fault tolerance
C.	greater throughput and enhanced fault tolerance
D.	none of the mentioned
Answer» D. none of the mentioned

Discussion

35.	The feature of multimicroprocessor architecture is
A.	task dependent
B.	single bus provider for many processors
C.	design is for a specific task
D.	all of the mentioned
Answer» E.

Discussion

36.	Memory management on a multiprocessor must deal with all of found on
A.	Uniprocessor Computer
B.	Computer
C.	Processor
D.	System
Answer» B. Computer

Discussion

37.	A multiprocessor operating system should perform
A.	a mechanism to split a task into concurrent subtasks
B.	optimise the system performance
C.	handling structural or architectural changes
D.	all of the mentioned
Answer» E.

Discussion

38.	A processor that continuously tries to acquire the locks, spinning around a loop till it reaches its success, is known as
A.	Spin locks
B.	Store locks
C.	Link locks
D.	Store operational
Answer» B. Store locks

Discussion

39.	An interface between the user or an application program, and the system resources is
A.	microprocessor
B.	microcontroller
C.	multimicroprocessor
D.	operating system
Answer» E.

Discussion

40.	The straight-forward model used for the memory consistency, is called
A.	Sequential consistency
B.	Random consistency
C.	Remote node
D.	Host node
Answer» B. Random consistency

Discussion

41.	If no node having a copy of a cache block, this technique is known as
A.	Uniform memory access
B.	Cached
C.	Un-cached
D.	Commit
Answer» D. Commit

Discussion

Explore topic-wise MCQs in Computer Architecture.

The directory must be tracking the group of nodes which have a copy of the block; then the used set is called

The particular block's statuses of physical memory are normally kept in one location, called

The pair of instructions including a special load known as a load-linked or load-locked with a special store called a

In the coherent multiprocessor, caches that is present provides both the migration and

Cases where variables gets updated without the order of the synchronization are called

One assigned operation for building synchronized operations, is called the

All writing procedures for same location are seen having same order; the stated property is called

For achieving a speedup up to 80 with 100 processors. What original computation fraction can be sequential

One that programming model which allows for having a more efficient implementation, is to suppose that programs are

Multiple-applications independently running, are typically called

The requesting node sending the requested data starting from the memory, and the requestor which has made the only sharing node, known as

Server clusters grow to 10th of thousands and beyond, are called

The tightly coupled set of threads' execution working on a single task, that is called

Microprocessors which are directly connected memory to a single-chip, that is sometimes called as

From inter-processor communication, the misses arises are often called

When every cache hierarchy level is a subset of level which is further away from the processor, is refered to as

The node which has the memory location and the entry of directory of an address is

Relaxing the W?R ordering, will yield a model known as

An operation being done in a way that intervening operation can be occurred, this operation is called a

Fetch-and-increment is another task of the

The protocols for maintaining coherence of multiple processors are known as

Private data that is used by a single-processor, then shared data are used by

Relaxing the W?W ordering, will yield a model known as

To update the cached copies of the data item; is the alternative protocol which is known as

The term which is named as shared memory with both SMP and DSM referring that the address space is

Symmetric multiprocessors architectures, are sometimes known as

Two-way set associative having a 64-byte block, the single clock-cycle hit time is a

When the home node being the local node, the copies may exist in a third node, called

A remote node is being the node which has a copy of a

The alternative design technique consists of multiprocessors having physically distributed memory, called

Only one node having a cache block copy, and this cache has written the block, and this memory copy is out of date. Then the processor is called the

Plotting performance versus number of processors, is refered to as

The alternative way of a snooping-based coherence protocol, is called a

The main objective in building the multimicroprocessor is

The feature of multimicroprocessor architecture is

Memory management on a multiprocessor must deal with all of found on

A multiprocessor operating system should perform

A processor that continuously tries to acquire the locks, spinning around a loop till it reaches its success, is known as

An interface between the user or an application program, and the system resources is

The straight-forward model used for the memory consistency, is called

If no node having a copy of a cache block, this technique is known as