Chapter 09 - Alternative Architectures Flashcards
(157 cards)
1
Q
What does RISC stand for?
A
Reduced Instruction Set Computer
2
Q
What does CISC stand for?
A
Complex Instruction Set Computer
3
Q
True or False: RISC architectures typically use a large number of instructions.
A
False
4
Q
True or False: CISC architectures are designed to execute complex instructions in a single cycle.
A
False
5
Q
Fill in the blank: RISC architectures focus on a ______ instruction set.
A
small
6
Q
Fill in the blank: CISC architectures aim to reduce the number of ______ required to execute a program.
A
instructions
7
Q
What is a key advantage of RISC architecture?
A
Higher performance due to simpler instructions and pipelining.
8
Q
What is a key disadvantage of CISC architecture?
A
Increased complexity and longer execution time for some instructions.
9
Q
Which architecture typically has a larger number of addressing modes, RISC or CISC?
A
CISC
10
Q
What is pipelining in the context of RISC?
A
A technique that allows overlapping execution of instructions to improve performance.
11
Q
Which architecture is generally more suited for applications requiring high performance and efficiency?
A
RISC
12
Q
How does instruction length typically compare between RISC and CISC?
A
RISC has fixed-length instructions, while CISC has variable-length instructions.
13
Q
True or False: RISC processors often require more lines of code compared to CISC processors for the same functionality.
A
True
14
Q
What is an example of a RISC architecture?
A
ARM or MIPS
15
Q
What is an example of a CISC architecture?
A
x86
16
Q
Which architecture typically has a more complex control unit?
A
CISC
17
Q
Fill in the blank: RISC architectures emphasize ______ execution of instructions.
A
fast
18
Q
What is the primary design philosophy behind CISC architectures?
A
To complete tasks with fewer instructions.
19
Q
True or False: RISC architectures minimize the number of cycles per instruction.
A
True
20
Q
Which architecture is more likely to use microcode?
A
CISC
21
Q
What is the main goal of RISC architecture?
A
To simplify instructions to enhance performance.
22
Q
Fill in the blank: CISC architectures can execute ______ instructions in a single operation.
A
complex
23
Q
What type of applications typically benefit from CISC architectures?
A
Applications that require rich instruction sets and complex operations.
24
Q
True or False: RISC processors are generally more power-efficient than CISC processors.
A
True
25
What does the term 'instruction throughput' refer to?
The number of instructions that can be executed in a given time period.
26
Which architecture allows for easier compiler optimization?
RISC
27
28
What is Flynn's Taxonomy?
A classification system for computer architectures based on instruction and data streams.
29
What are the four categories in Flynn's Taxonomy?
SISD, SIMD, MISD, and MIMD.
30
True or False: SISD stands for Single Instruction Single Data.
True.
31
Fill in the blank: SIMD stands for _______.
Single Instruction Multiple Data.
32
What does MISD stand for?
Multiple Instruction Single Data.
33
Which category of Flynn's Taxonomy is characterized by multiple processors executing different instructions on the same data?
MIMD.
34
True or False: SIMD architectures are ideal for tasks that can be parallelized.
True.
35
What type of applications typically benefit from MIMD architectures?
Applications requiring complex and diverse computations.
36
Which Flynn's Taxonomy category is the most common in modern computers?
SISD.
37
Multiple choice: Which of the following is an example of SIMD?
Graphics processing.
38
What is one key advantage of MIMD architectures?
They can handle multiple tasks simultaneously.
39
Fill in the blank: In MISD, multiple instructions operate on a _______.
single data stream.
40
True or False: Flynn's Taxonomy only applies to hardware architectures.
False.
41
What is a primary disadvantage of SISD architectures?
Limited performance for parallel processing tasks.
42
Multiple choice: Which of the following best describes SIMD?
One instruction operating on multiple data points.
43
What is the main characteristic of MISD architectures?
They have multiple instruction streams but a single data stream.
44
Fill in the blank: MIMD systems can be used for both _______ and _______ tasks.
parallel; sequential.
45
True or False: SIMD can be used effectively for image processing.
True.
46
In Flynn's Taxonomy, what does SISD typically represent?
A traditional single-core CPU.
47
What is one example of a system using MIMD?
Multi-core processors.
48
Multiple choice: Which Flynn's category is least commonly used?
MISD.
49
True or False: Flynn's Taxonomy helps in understanding parallel computing.
True.
50
What type of computing does SIMD excel in?
Data-parallel computing.
51
Fill in the blank: In MIMD, each processor can execute _______ instructions independently.
different.
52
What is the primary focus of Flynn's Taxonomy?
The relationship between instruction and data streams in architectures.
53
True or False: Flynn's Taxonomy is relevant only for theoretical computer science.
False.
54
What is the significance of Flynn's Taxonomy in modern computing?
It provides a framework for understanding and designing parallel processing systems.
55
56
What does MIMD stand for?
Multiple Instruction, Multiple Data
57
True or False: MIMD systems can execute different instructions on different data simultaneously.
True
58
Which of the following is a subtype of MIMD? A) SIMD B) SPMD C) MISD D) None of the above
B) SPMD
59
Fill in the blank: MIMD systems are characterized by ____ execution of instructions.
independent
60
What is a key advantage of MIMD architectures?
Flexibility in processing different tasks
61
Name one application where MIMD architectures are commonly used.
High-performance computing
62
True or False: MIMD systems require complex control mechanisms.
True
63
What distinguishes MIMD from SIMD?
MIMD can execute multiple instructions on multiple data streams, while SIMD executes the same instruction on multiple data points.
64
Fill in the blank: In MIMD systems, each processor can operate on ____ data streams.
different
65
What is the primary challenge of MIMD architectures?
Synchronization and communication between processors
66
Name a common MIMD architecture type.
Cluster computing
67
True or False: MIMD systems are less efficient than SIMD systems for parallel processing tasks.
False
68
What is the full form of SPMD in the context of MIMD?
Single Program, Multiple Data
69
Which type of MIMD architecture allows different processes to run different programs?
Distributed MIMD
70
True or False: MIMD is suitable for both tightly and loosely coupled systems.
True
71
What is the role of a scheduler in MIMD systems?
To manage the execution of different instructions on multiple processors
72
Fill in the blank: MIMD systems can be implemented using ____ and ____ processors.
homogeneous; heterogeneous
73
What does MISD stand for in relation to MIMD?
Multiple Instruction, Single Data
74
True or False: MIMD systems are ideal for real-time processing tasks.
False
75
Which of the following is NOT a characteristic of MIMD systems? A) Independent execution B) Different instructions C) Single memory space D) Multiple data streams
C) Single memory space
76
What is one disadvantage of MIMD architectures?
Higher complexity in programming and management
77
In MIMD, what does the term 'task parallelism' refer to?
The ability to execute different tasks simultaneously across processors
78
Fill in the blank: MIMD systems are often used in ____ computing environments.
cloud
79
What is a hybrid approach in MIMD systems?
Combining MIMD with SIMD to optimize performance
80
True or False: MIMD can improve resource utilization in multi-core processors.
True
81
What is the main difference between MIMD and MPMD?
MPMD refers to Multiple Program, Multiple Data, where multiple programs run concurrently on multiple data streams.
82
What does MIMD stand for?
Multiple Instruction Multiple Data
83
True or False: MIMD allows multiple processors to execute different instructions on different data.
True
84
What is the main characteristic of NOW in MIMD?
Network of Workstations
85
Fill in the blank: COW stands for _______.
Cluster of Workstations
86
Which MIMD model is known for being a tightly integrated system?
DCPC (Distributed Cache Parallel Computer)
87
What does POPC stand for?
Parallel Object-oriented Programming for Clusters
88
True or False: COW is primarily used for high-performance computing tasks.
True
89
What type of architecture does NOW utilize?
Loose coupling of individual workstations
90
Which MIMD architecture is typically easier to scale?
COW
91
What is a primary advantage of using DCPC?
Efficient data sharing among processors
92
True or False: POPC is specifically designed for single-threaded applications.
False
93
What is a key feature of NOW?
Collaboration of multiple independent systems
94
Which MIMD approach focuses on the utilization of object-oriented programming?
POPC
95
Fill in the blank: DCPC systems often utilize _______ for data management.
Distributed caches
96
What is the primary use case for COW architectures?
Running parallel applications on clusters
97
True or False: NOW systems are generally less expensive than traditional supercomputers.
True
98
What is a disadvantage of using NOW?
Potential for network bottlenecks
99
Which architecture allows for dynamic resource allocation?
COW
100
What programming model is typically used in POPC?
Object-oriented programming model
101
Fill in the blank: MIMD systems can be categorized into _______ and _______ types.
Shared memory; Distributed memory
102
Which MIMD model is best for applications requiring high inter-process communication?
DCPC
103
True or False: The NOW architecture supports heterogeneous computing environments.
True
104
What is a common challenge faced by COW systems?
Load balancing among workstations
105
What is the primary goal of MIMD architectures?
To enhance parallel processing capabilities
106
Fill in the blank: In MIMD, each processor can execute _______ instructions independently.
Different
107
Which MIMD architecture is typically more cost-effective for small to medium-sized tasks?
NOW
108
What is pipelining in computer architecture?
Pipelining is a technique where multiple instruction phases are overlapped to improve CPU instruction throughput.
109
True or False: Pipelining increases the number of instructions that can be executed per clock cycle.
True
110
Fill in the blank: In a pipelined processor, the stages of instruction execution are divided into _____ stages.
multiple
111
What are the typical stages in a basic instruction pipeline?
Fetch, Decode, Execute, Memory Access, Write Back
112
What is superpipelining?
Superpipelining is an extension of pipelining that increases the number of pipeline stages to allow for higher clock speeds.
113
Multiple choice: Which of the following is a benefit of superpipelining?
A) Reduced complexity
B) Higher throughput
C) Lower power consumption
B) Higher throughput
114
What is the main advantage of superscalar architectures?
Superscalar architectures can issue multiple instructions in a single clock cycle.
115
True or False: Superscalar architectures require multiple execution units.
True
116
What is a key challenge in implementing superscalar architectures?
Instruction-level parallelism and managing dependencies between instructions.
117
Fill in the blank: In superscalar architectures, the ability to execute multiple instructions simultaneously is called _____.
instruction-level parallelism
118
What is the term for the situation when two instructions depend on the same resource?
Resource contention
119
Multiple choice: Which of the following is NOT a characteristic of pipelined processors?
A) Overlapping instruction execution
B) Increased instruction latency
C) Reduced cycle time
B) Increased instruction latency
120
What does the term 'hazard' refer to in pipelining?
A condition that prevents the next instruction in the pipeline from executing.
121
True or False: Data hazards can occur when an instruction depends on the result of a previous instruction.
True
122
Fill in the blank: Structural hazards occur when hardware resources are _____ by multiple instructions.
contended
123
What is the main difference between superscalar and superpipelined architectures?
Superscalar architectures focus on executing multiple instructions per cycle, while superpipelined architectures focus on increasing the number of pipeline stages.
124
Multiple choice: Which of the following techniques is used to resolve data hazards?
A) Instruction reordering
B) Branch prediction
C) Pipelining
A) Instruction reordering
125
What is a branch hazard?
A situation in pipelining where the next instruction to execute is uncertain due to a branch instruction.
126
True or False: Superscalar processors can have multiple instruction decode units.
True
127
Fill in the blank: The process of breaking down a complex instruction into simpler operations is known as _____.
instruction decomposition
128
What is out-of-order execution?
A technique that allows instructions to be executed as resources are available, rather than strictly in the order they appear.
129
Multiple choice: Which of the following best describes a pipeline stall?
A) A pause in instruction execution
B) An increase in clock speed
C) A decrease in instruction throughput
A) A pause in instruction execution
130
What is the purpose of a scoreboard in superscalar architectures?
To track the status of instructions and their dependencies to optimize execution.
131
True or False: Superpipelining typically results in greater complexity compared to basic pipelining.
True
132
What does VLIW stand for?
Very Long Instruction Word
133
True or False: VLIW architectures allow multiple operations to be executed in parallel.
True
134
What is the main advantage of VLIW architectures over traditional architectures?
Increased instruction-level parallelism
135
Fill in the blank: VLIW architectures rely on the compiler to schedule instructions for _______.
parallel execution
136
Which of the following is a key component of VLIW architectures? A) Control Unit B) Instruction Decoder C) Instruction Bundling
C) Instruction Bundling
137
What type of computers are designed to handle large data sets and perform operations on vectors?
Vector Computers
138
True or False: Vector computers process data in a scalar manner.
False
139
What is a primary characteristic of vector processing?
Operating on one-dimensional arrays of data
140
In VLIW architectures, who is responsible for instruction scheduling?
The compiler
141
Fill in the blank: Vector computers typically utilize _______ to achieve high throughput.
pipelining
142
What is the purpose of instruction bundling in VLIW architectures?
To group multiple instructions into a single long instruction word
143
Which architecture type typically requires more complex compilers, VLIW or RISC?
VLIW
144
True or False: Vector computers can efficiently handle tasks such as matrix operations.
True
145
What is the key benefit of using vector processors for numerical computations?
They can perform the same operation on multiple data points simultaneously.
146
What is one challenge associated with VLIW architectures?
Code size may increase due to instruction bundling.
147
Fill in the blank: The main processing unit of a vector computer is called a _______.
vector processor
148
What is the primary difference between scalar and vector processing?
Scalar processes one data item at a time, while vector processes multiple data items simultaneously.
149
Which type of computer architecture is known for its ability to perform SIMD operations?
Vector Computers
150
True or False: VLIW architectures require hardware support for dynamic scheduling.
False
151
What does SIMD stand for in the context of vector processing?
Single Instruction, Multiple Data
152
What is a potential drawback of VLIW architectures regarding instruction execution?
Under-utilization of resources if there are not enough independent instructions.
153
Fill in the blank: Vector architectures often use _______ memory to store vector data.
contiguous
154
What is a common application area for vector computers?
Scientific computing and simulations
155
True or False: VLIW architectures can reduce the need for complex hardware.
True
156
What is one reason why VLIW architectures may not be widely adopted?
The reliance on compilers for optimization can be limiting.
157
What is the term for the grouping of instructions in VLIW architectures?
Instruction bundling
