ENG210 Computer Architecture: CPU & Memory

0.0(0)
studied byStudied by 0 people
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
Card Sorting

1/113

flashcard set

Earn XP

Description and Tags

A comprehensive set of 250 flashcards covering key concepts of Computer Architecture, specifically on CPU and memory components, processes, and enhancements.

Study Analytics
Name
Mastery
Learn
Test
Matching
Spaced

No study sessions yet.

114 Terms

1
New cards

What are the major components of the CPU?

ALU (Arithmetic Logic Unit), CU (Control Unit), Memory Management Unit, and I/O Interface.

2
New cards

What does the ALU perform?

Calculations, comparisons, and executes logical operations (AND, OR, NOT, XOR).

3
New cards

What is the role of the Control Unit (CU)?

It performs the fetch/execute cycle, accesses program instructions, issues commands to the ALU, and manages data movement.

4
New cards

What does the Memory Management Unit do?

It supervises the fetching of instructions and data from memory.

5
New cards

What are registers in the CPU?

Small, permanent storage locations with specific functions that hold data, addresses, or instructions.

6
New cards

What is the function of General Purpose Registers?

They hold intermediate results or data values and are visible to the user or program.

7
New cards

What is the Program Count Register (PC) also known as?

Instruction Pointer.

8
New cards

What does the Instruction Register (IR) do?

Stores the instruction fetched from memory.

9
New cards

What is the function of the Memory Address Register (MAR)?

Holds the address of a memory location to be accessed.

10
New cards

What does the Memory Data Register (MDR) hold?

Temporarily holds data that is being transferred to or from memory.

11
New cards

Define volatile memory.

Memory that requires constant power to maintain the stored information, like DRAM.

12
New cards

What does RAM stand for?

Random Access Memory.

13
New cards

Describe Dynamic RAM (DRAM).

It is the most common type of RAM, cheaper, less power-consuming, volatile, and needs to be refreshed often.

14
New cards

What is Static RAM (SRAM)?

Faster and more expensive than DRAM and is often used as cache memory.

15
New cards

What is ROM?

Read-Only Memory, which holds software expected to remain unchanged.

16
New cards

What is EEPROM?

Electrically Erasable Programmable Read-Only Memory.

17
New cards

What does cache memory do?

Stores frequently accessed data for quicker retrieval, improving CPU efficiency.

18
New cards

Explain the Fetch-Execute Cycle.

A two-step process where the CPU fetches a program instruction from memory and executes it.

19
New cards

What happens during the Fetch Phase?

The address from the Program Counter (PC) is placed in the MAR, the instruction is loaded into IR, and the PC is incremented.

20
New cards

What memory types are nonvolatile?

ROM, EEPROM, and Flash Memory.

21
New cards

What is a bus in computer architecture?

A physical connection that transfers data from one location to another.

22
New cards

What are the four types of bus signals?

Data, Addressing, Control signals, and Power.

23
New cards

What is pipelining in CPU architecture?

An assembly-line technique that allows overlapping fetch-execute cycles for multiple instructions.

24
New cards

Define scalar processing.

Average instruction execution is approximately equal to the CPU clock speed.

25
New cards

What is superscalar processing?

The ability of a CPU to execute more than one instruction per clock cycle.

26
New cards

Explain the purpose of the Fetch Unit.

Handles instruction fetching and decoding from memory.

27
New cards

Describe the Execution Unit's role.

Executes the instructions received from the decode unit.

28
New cards

What are the advantages of cache memory?

Higher hit ratio leading to improved execution speed due to locality of reference.

29
New cards

Define multiprocessing in computer architecture.

Using two or more CPUs to increase processing power.

30
New cards

What is master-slave multiprocessing?

One master CPU manages the system and assigns tasks to slave CPUs.

31
New cards

Define symmetrical multiprocessing (SMP).

Each CPU has equal access to resources and can run processes independently.

32
New cards

What are the drawbacks of master-slave multiprocessing?

The master CPU can become a bottleneck and failure can lead to complete system failure.

33
New cards

What is the significance of performance advantages from cache memory?

It allows for faster data access, minimizing the time the CPU waits for data retrieval.

34
New cards

What causes latency issues in memory access?

Differences in speed between the CPU processes and memory access speeds.

35
New cards

What types of operations do CPUs perform?

Data movement, arithmetic, logic operations, and I/O machine control.

36
New cards

What is the function of the instruction set?

Defines the functions performed by the processor and differentiates computer architectures.

37
New cards

Give an example of a special-purpose register.

Program Count Register (PC) or Instruction Register (IR).

38
New cards

How does the Control Unit interact with the ALU?

It delivers data to the ALU and informs it which operation to perform.

39
New cards

What is a hit ratio in cache memory?

The ratio of successful data retrievals from cache to total requests.

40
New cards

What is meant by instruction reordering?

The rearrangement of instruction execution to avoid delays due to dependencies.

41
New cards

Explain memory interleaving.

Partitioning memory into subsections for faster access through simultaneous addressing.

42
New cards

How does a CPU know which instruction to execute next?

By using the Program Counter (PC) to point to the next instruction address.

43
New cards

What is meant by 'out-of-order processing'?

Executing instructions not strictly in the order they appear, to optimize pipeline performance.

44
New cards

What does the CPU's clock speed determine?

The rate at which the CPU can execute instructions.

45
New cards

What happens when there is a cache miss?

The cache controller must fetch the required data from the main memory.

46
New cards

Define the term 'bus throughput.'

The data transfer rate in bits per second for a bus.

47
New cards

What is the role of the cache controller?

It manages the flow of data between the cache and main memory.

48
New cards

How does the ALU support comparisons?

By executing logical operations to evaluate conditions like greater than or equal.

49
New cards

What does pipelining help improve in CPU performance?

It reduces the execution time of sequences of instructions by overlapping cycles.

50
New cards

What is the impact of data dependencies in CPU execution?

They can stall the instruction pipeline, affecting processing speed.

51
New cards

How is instruction format significant in computer architecture?

It defines the layout of the instructions, including opcode and operand locations.

52
New cards

What is the difference between a fixed and variable instruction size?

Fixed size means all instructions have the same length; variable size can vary.

53
New cards

What do flags in the CPU indicate?

They track conditions like overflow, carry, and errors during operations.

54
New cards

How many instruction types does a CPU generally support?

Multiple types including data movement, arithmetic, and control instructions.

55
New cards

What is a bus interface unit?

A component that manages data exchanges between the CPU and memory or peripherals.

56
New cards

What is the purpose of a status register?

To hold information about the current state of operations in the CPU.

57
New cards

What does 'locality of reference' mean in caching?

The tendency of a processor to access a small set of memory locations frequently.

58
New cards

What is branch prediction?

A technique to guess the outcome of a branching operation to improve pipeline efficiency.

59
New cards

What is the function of an instruction pointer?

To track the address of the next instruction to be executed.

60
New cards

What are point-to-point buses?

Connections that link a single source to a single destination.

61
New cards

Define a multipoint bus.

A bus that connects multiple devices to a single communication channel.

62
New cards

What is the primary purpose of cache lines?

To serve as a unit of transfer between storage and cache memory.

63
New cards

How does the memory bus affect data transfer?

It impacts the speed and bandwidth for data communication in the system.

64
New cards

What is the role of a cache line in cache memory?

To organize data retrieved from main memory for better efficiency.

65
New cards

What setting allows cache memory to deliver better performance?

A higher hit ratio and frequent data locality within processes.

66
New cards

Why is the CPU unable to access memory directly?

The bus and cache memory act as intermediaries to facilitate access to RAM.

67
New cards

What are common application outcomes of high-speed memory access?

Faster program execution and improved overall system performance.

68
New cards

What can cause CPU bottlenecking in a master-slave system?

The master CPU being overwhelmed with too many requests from slave CPUs.

69
New cards

What constitutes an effective architecture in multiprocessor systems?

Balanced distribution of processes and efficient resource sharing.

70
New cards

What is the functional difference between pipeline execution and superscalar processing?

Pipelining overlaps execution phases, while superscalar executes multiple instructions simultaneously.

71
New cards

Define the term 'observation of bottlenecks' regarding CPU usage.

Identifying moments where processing efficiency drops due to resource limitations.

72
New cards

How does instruction size affect pipeline architecture?

Larger and varied sizes complicate simultaneous instruction processing.

73
New cards

What does a 'fetch signal' do during memory accesses?

It prompts the memory to send a specific piece of data requested by the CPU.

74
New cards

What is the impact of using different clock signals for CPUs?

It can lead to synchronization issues and performance inconsistencies.

75
New cards

What defines cache's hit consistency across multiple processes?

Sharing a unified cache memory to maintain state across multiple cores.

76
New cards

How is addressability of a system determined?

By the number of bits in the MAR and the span of addressable locations.

77
New cards

What does a sample CPU organization look like?

It includes an execution unit, fetch unit, memory bus, and cache.

78
New cards

Explain the working of the Cache Write Back strategy.

Data is written back to the main memory only when necessary, improving write speeds.

79
New cards

What is the functionality of Address Decoders in relation to memory?

They translate logical addresses to physical addresses for data access.

80
New cards

How does computer architecture distinguish between various CPU designs?

Through differences in instruction sets, execution speeds, and parallel processing capabilities.

81
New cards

Why is managing CPU registers crucial for performance?

To optimize the speed of data retrieval and processing within the CPU.

82
New cards

Define the significance of Bus Characteristics in system performance.

They dictate the efficiency of data throughput and control mechanisms.

83
New cards

What is meant by 'Arithmetic Logic Unit' in processor design?

The component responsible for performing mathematical and logical operations.

84
New cards

Explain how online performance monitoring can alleviate bottlenecks.

By adjusting workloads dynamically based on real-time CPU usage.

85
New cards

What are power reduction features in modern CPUs?

Techniques such as dynamic frequency scaling and sleep modes to conserve energy.

86
New cards

Identify the role of Conditional Branching in CPU operation.

To enable decisions in program flow based on specific boolean evaluations.

87
New cards

What are the challenges posed by data dependencies?

They can hinder pipeline execution efficiency and require careful management.

88
New cards

Define cache coherence in multiprocessor systems.

Ensuring that multiple caches have a consistent state regarding shared data.

89
New cards

How does a computer classify immediate and indirect addressing modes?

Immediate uses explicit values, while indirect fetches values from a location.

90
New cards

What is the impact of instruction set architecture on software development?

It influences the efficiency and complexity of programming needed for optimal performance.

91
New cards

What does the term 'hazard detection' refer to in pipelining?

Identifying conditions that may cause pipeline stalls or incorrect execution.

92
New cards

Explain why different CPU designs use varying instruction lengths.

To balance complexity with efficiency based on computational requirements.

93
New cards

How does clock rate influence CPU performance?

Higher clock speeds allow the execution of more instructions in a given timeframe.

94
New cards

What does changing cache sizes do to access times?

Larger caches may speed up access but can increase retrieval times if they're not well organized.

95
New cards

What types of multiprocessor systems are generally designed for high parallelism?

Symmetrical multiprocessing systems that allow equal resource access for all processors.

96
New cards

Describe the impact of instruction scheduling on pipeline performance.

Proper scheduling can enhance throughput and minimize stalls in the pipeline.

97
New cards

What challenges exist in between different CPUs sharing resources?

Resource contention can lead to delays and bottlenecks affecting overall system performance.

98
New cards

How can branch prediction affect CPU cycles?

By anticipating branching decisions, it reduces stalls and keeps the pipeline full.

99
New cards

What role does instruction scheduling play in reducing the execution time?

It organizes instructions to maximize parallel execution and minimize CPU idle time.

100
New cards

How are memory addresses structured in programs?

As sequential locations determined by the addressing scheme of the CPU.