Chapter 9 test Flashcards
(116 cards)
1
Q
A physical memory cache is designed to search the local cache or access the underlying memory, but not in parallel.
A
False
2
Q
A reference to a missing page is called a page fault.
A
T
3
Q
A virtual address space can be contiguous.
A
T
4
Q
After separating read and write operations, there is no need for additional measurement.
A
???
5
Q
Although a variety of caching mechanisms exist, they share 4 general characteristics. Name two of the four characteristics.
A
Small, Active, Transparent, Automatic
6
Q
An alternate to segmentation that has been very successful is known as demand padding.
A
T
7
Q
An architect views a memory as a solid-state digital device that provides storage for data values.
A
T
8
Q
An MMU creates a real address for the processor.
A
F
9
Q
Any measure of memory performance must give two values. What operations do those values represent?
A
Read operations and wrote operations.
10
Q
Byte addressing is convenient for programming because it gives a programmer an easy way to access small data items.
A
T
11
Q
Cache designers use the term hit miss ration to refer to the percentage of requests in the sequence that are not satisfied from the cache.
A
???
12
Q
Caching can be used with hardware and software.
A
T
13
Q
Caching cannot be used with non-textual data.
A
F
14
Q
Caching has 4 general characteristics: it is large, static, visible, and not automatic.
A
F
15
Q
Caching reduces the Von Neumann bottleneck
A
T
16
Q
Computers that use DRAM contain an extra hardware circuit called static circuit.
A
F
17
Q
Conceptually, SRAM stores each data bit in a miniature digital circuit composed of multiple transistors similar to the flip-flop.
A
T
18
Q
Correct terminology when speaking about page frames is that software loads a page into memory of frame.
A
T
19
Q
Currently, the distinction between L1 and L2 cache is fading.
A
T
20
Q
Data items in a cache are not limited to a specific type, form, or size.
A
T
21
Q
Density and latency stand out when referring to access memory technology.
A
T
22
Q
Despite the cost and power consumption advantages of DRAM most computer memory is composed of SRAM rather than DRAM.
A
F
23
Q
Dividing a virtual address space on a boundary that corresponds to a power of two allows the MMU to perform address translations without requiring arithmetic operations.
A
T
24
Q
Early computer designers thought that multi-programming was impractical.
A
T
25
Engineers use this term to denote the type of memory used as the primary memory system in most computers.
Random access memory.
26
For virtual memory schemes, no single scheme is optimal in all cases.
T
27
Given a parallel connection between a processor and memory, a connection that contains N wires, allows 2^N bits of data to be transferred simultaneously.
F
28
How do architects view memory?
A solid-state digital device that provides storage for data values.
29
How is physcial memory organized?
It is organized into words.
30
In essence, a DRAM chip is a perfect memory device.
F
31
In order for a cache to resolve the ambiguity that occurs because multiple applications use the same range of address two solutions have been created. Name one of th esolutions.
Cache flush operations, A disambiguating identifier.
32
In paging, the number of bytes per page is chosen to be a power of 8.
F
33
In physical memory each read and write operation applies to an entire world.
T
34
In segmentation, each program is divided into fixed-sized blocks called pages.
F
35
It is not possible for a memory management unit to use two different types of physical memory in a single virtal address space.
F
36
Many processors use bit addressing because it provides the most convenient interface for programmers.
T
37
Memory organization refers to two structures. What are the two structures and label them as internal or external?
Internal hardware structure and external addressing structure.
38
Memory systems use a ___ that controls exactly when a read or write operations begins.
Clock
39
Memory systems use a clock that controls exactly when a read or write operation begins.
T
40
MMU cannot use what operation to implement address translation.
Subtraction
41
Most personal computers use a cache hierarchy. The cache at the top of the hierarchy is usually the smallest and fastest of the cache hierarchy.
T
42
Most ram is volatile.
T
43
Name the two different types of RAM technologies.
SRAM and DRAM
44
Name the two technologies needed for a virtual memory system that supports demand paging.
Hardware that handles address mapping and software that moves pages between.
45
Name two characteristics of caching mechanisms.
Small and active.
46
N-Way interleaving describes the theoretical number of underlying memory modules.
F
47
Once the cache storage is full what must it use to decide how to handle further new items?
Replacement policy
48
One technology is needed for a virtual memory system that supports demand paging.
F
49
Memory used to hold a program is used as instruction store and memory used to hold data is known as data store.
T
50
Random Access Memory denotes the secondary memory system rarely used in computers.
F
51
Random Access requires that values must be read from memory in the same order they were inserted.
F
52
Real address space refers to the set of addresses recognized by the virtual memory
F
53
Segmentation faded from popularity because it led to a problem called fragmentation.
T
54
Segmentation refers to a virtual memory scheme in which programs are divided into variable size blocks, and only the blocks currently needed are kept in memory. Segmentation often leads to fragmentation and is often used today.
F
55
Some memory systems use a synchronized clock system to align the clock pulses used by the memory system and the processor.
T
56
SRAM stores each data bit in a miniature digital circuit composed of multiple transistors.
T
57
The hit ratio is extremely low during startup. What is a method to improve the hit ratio specifically during startup?
Preloading the cache
58
The Last Recently Used (LRU) cache replacement system uses a policy that specifies replacing the item that was used the longest time in the past.
F
59
The MMU receives memory requests from the processor.
T
60
The most common forms of memory are classified as selected access.
F
61
The most common technology for multiprogramming uses physical memory to establish a separate virtual address space for each program.
F
62
The organizaiton of physical memory affects programming.
T
63
The processor begins running in real mode, which means that the processor passes all memory references to the MMU.
F
64
The purpose of page replacement is to reduce the overhead of creating a new page in memory.
F
65
The term caching refers to an important optimization technique used to reduce the Von Neumann bottleneck.
T
66
The term frame refers to a block of a program's address space.
F
67
The three control bits found in most hardware are the presence bit, the past bit, and the modified bit.
F
68
There is only one key aspect of memory organization: size.
F
69
To avoid arithmetic calculations, such as division or remainder, physical memory is organized such that the number of bytes per word is a power of three. This means the translation from a byte address to a word address can be performed and offset by extracting the bits.
F
70
To implement uniform virtual address space out of two dissimilar physical memory systems, we use software to perform necessary translations.
F
71
Using parallelism some architectures provide higher memory performance using multiple memory banks.
T
72
We use the term virtual memory to refer to a mechanism that hides the details of underlying physical memory.
T
73
What are the three steps that are required to translate a virtual memory address V, to a corresponding physical address, P?
???
74
What are two disadvantages of SRAM
Power consumption and heat
75
What do architects use the term real to refer to?
Parts of a physical memory
76
What does a bus provide?
Parallel connecitons
77
What does a presence bit specify?
Determine whether page is currently present in memory.
78
What does an architect view as a solid-state digital device that provides storage for data values?
Memory
79
What does it mean for memory to be classified as volatile?
If the contents of the memory disappear when power is removed.
80
What does MMU stand for?
Memory Management Unit
81
What does the term density refer to?
The number of memory cells per square area of silicon.
82
What does the term page refer to?
A block of a program's address space.
83
What does the term volatile mean when referring to RAM?
Values do not persist after the computer is powered down.
84
What is a contiguous virtual address space?
An address space in which all addresses are mapped onto an underlying physical memory.
85
What is a mechanism that hides the details of the underlying physical memory to provide a more convenient memory environment?
Virtual Memory (VM)
86
What is a multi-level cache hierarchy?
When more than one cache is used along the path from requester to data store.
87
What is a pointer?
A variable that holds a memory address.
88
What is a popular variant of EEPROM technology?
Flash memory
89
What is a reference to a missing page called?
Page fault
90
What is a refresh circuit?
Performs the task of reading and then writing a bit.
91
what is a virutal memory technology that maps an entire address space called?
Page fault.
92
What is byte alignment?
An integer value is aligned if the bytes of the integer correspond to a word in the underlying physical memory.
93
What is a coarse granularity mapping?
Virtual memory technology that maps an entire address space.
94
What is interleaving?
Spreads consecutive bytes of memory across separate physical memory modules.
95
What is one of the four facts that caused computer architects to add a second level of cache to the memory hierarchy?
A traditional memory cache was separate from both the memory and the processor.
96
What is row major order?
Rows of an array are placed in contiguous memory.
97
What is segmentation?
Each program is divided into variable-sized blocks, and the pieces of program are kept on an external storage device until one is needed.
98
What is the central idea in caching?
High-speed storage.
99
What is the effect of different clock rates on RAM?
Performance.
100
What is the most popular replacement policy for replacing cache?
Least Recently Used (LRU)
101
What is the most significant limitation for the size of memory?
A processor imposes a fixed bound on the size of an address that can be generated.
102
What is the name of the small blocks of memory created by segmentation?
Segments
103
What is the significant disadvantage to SRAM?
Power consumption and heat
104
What is the technical name for the hardware connection between a processor and memory.
Bus
105
What is Translation Lookaside Buffer (TLB)?
Is used to optimize performance of a paging system.
106
What is virtual memory?
A mechanism that hides the details of the underlying physical memory to provide a more convenient memory environment.
107
What lower cost memory can be used to in place of SRAM?
DRAM
108
What mechanism is the easiest dynamic virtual memory scheme to understand.
???
109
What paradigm do all memories technologies use?
fetch-store
110
What two operations does the controller for physical memory support?
Read and write.
111
What would result if a processor attempted to read an address that does not correspond to physical memory?
An error.
112
When an architect begins to design a memory, what two keys choices arise?
Technology and organization
113
When its storage is full and a new item arrives, a cache must choose whether to retain the current set of items or save them to disk.
F
114
When speaking about demand paging to what does a page resident page refer?
When a page is in memory.
115
Where is a cache placed?
On the path between a mechanism that makes requests.
116
Why do Memory Management Units avoid using subtraction to implement address translation?
Because subtraction requires substantial hardware (ALU)
