Week 5 and 6

Question 1

What are the three different stages of Binding Instructions and Data

Answer 1

1. Compile (and linkage) time
2. Load Time - When loader loads program into memory, addresses are resolved
3. Execution Time - programs move in memory - hardware support required

Question 2

Def: Execution time

Answer 2

Similar to Compile Time, but hardware looks after translation

Question 3

Def: Logical Address

Answer 3

Address generated by CPU, also known as virtual address

Question 4

Def: Physical Address

Answer 4

location in physical memory

Question 5

What are the similarities and differences of Logical and Physical addresses ?

Answer 5

• are the same in compile and load time address binding
• differ in execution time binding

user program only deals with logical addresses (never sees physical)

Question 6

Def: MMU

Answer 6

Hardware that maps virtual to physical address

- one simple approach is to have a single register that is added to every virtual address

Question 7

What is contiguous allocation?

Answer 7

Main memory is divided into two parts:

1. Operating system
2. user memory (everything else)

with this approach each process is contained in a single contiguous section of memory

Question 8

What are the three general storage allocation approaches?

Answer 8

1. First Fit (use the first hold on the free list that is big enough, search from previous location/ only part of list)
2. Best fit (smallest block that is large enough, search entire list)
3. Worst Fit (largest block, worst algorithm)

Question 9

Def: Internal fragmentation

Answer 9

if allocate memory in units larger than single byte, last block only partially used (some memory no used because more allocated to process then needed)

Question 10

Def: external fragmentation

Answer 10

lots of small holes spread throughout memory, none big enough to satisfy a request (worst fit tries to reduce this, compaction - move blocks requires execution-time binding)

Question 11

What is the 50 percent rule? (to do with Fragmentation Trade off)

Answer 11

N allocated blocks, 0.5 N lost to fragmentation (1/3 of memory unusable) `

Question 12

What are the two Fragmentation Trade offs?

Answer 12

Allocate in smaller unit - less internal fragmentation, overhead in managing)
Allocate in larger units - easier to manage, more internal fragmentation

Question 13

Def: Dynamic Loading

Answer 13

routines are loaded with needed (not necessarily when program loaded)
Does not require any special support from operating system

Question 14

Def: Static linking

Answer 14

when all of the modules including system libraries are linked together at compile time

Question 15

Def: Dynamic linking

Answer 15

uses indirect call through table
- table initially points at stubs
- when the routine is called the first time, the routing is leaded
(primarily used for common libraries)

Question 16

Def: Overlays

Answer 16

program is broken into multiple parts - one common part of program always in memory - other parts of program are replaced as needed

Question 17

Def: Swapping

Answer 17

Processes can be temporarily stored (swapped) from memory to a backing store
- make room for higher priority processes

Question 18

Describe paging.

Answer 18

• Phys. mem. divided into frames (512 bytes to 8K sizes typically)
• Logical memory is divided into pages (same size as frames)
• if process needs n pages, find n free frames in memory
• table translates from each page to the appropriate frame
• no external fragmentation (still internal)

Question 19

What is a page tables ?

Answer 19

A page table is the data structure used by virtual memory system in a computer operating system to store the mapping between virtual addresses and physical addresses
- each process has its own

Question 20

What is a frame table?

Answer 20

Keeps track of allocated and free frames

Question 21

Def: TLB

Answer 21

Translation look-aside buffer

- a memory cache that is used to reduce the time taken to access a user memory location (part of the the MMU)

Question 22

What are the three ways to reduce the memory requirements of the page tables?

Answer 22

1. Hierarchical Tables (multiple tables -> forward-mapped page table, page number is split into one or more sections)
2. Hashed Page Tables (store page entries in a hash table, hashed by page number)
3. Inverted pages (one entry for each frame - unlike other two which have separate tables per process)

Question 23

What is virtual memory?

Answer 23

Separation of Logical address from Physical Address

Virtual memory is a memory management capability that uses hardware and software to allow a computer to compensate for physical memory shortages by temporarily transferring data from random access memory (RAM) to storage

Question 24

What is Demand Paging?

Answer 24

• All pages of the process are swapped out

- swap each page in as it is needed

Question 25

Locality of reference

Answer 25

same values , or related storage locations, are frequently accessed depending on the memory access pattern

Question 26

Def: memory resident

Answer 26

Pages that are in memory

Question 27

What is the Traditional I/O for memory mapped files?

Answer 27

Library accumulates until a buffer is filled and then calls O/S to write the buffer to the fule

Question 28

What is over allocation?

Answer 28

running more processes than memory

Question 29

What is the dirty bit?

Answer 29

Indicates that the page has been changed (dirty)