Chapter 9: Virtual Memory

Chapter 9: Virtual
Memory



Background
Demand Paging
 Copy-on-Write
 Page Replacement
 Allocation of Frames
 Thrashing
 Memory-Mapped Files
 Allocating Kernel Memory
 Other Considerations
 Operating-System Examples
Objectives

 To describe the benefits of a virtual memory

system

 To explain the concepts of demand paging, page-

replacement algorithms, and allocation of page
frames

 To discuss the principle of the working-set model

Background
 Virtual memory – separation of user logical memory from
physical memory.
 Only part of the program needs to be in memory for execution
 Logical address space can therefore be much larger than physical
address space
 Allows address spaces to be shared by several processes
 Allows for more efficient process creation

 Virtual memory can be implemented via:

 Demand paging
 Demand segmentation
Virtual Memory That is Larger Than Physical
Memory


Virtual-address Space
Shared Library Using
Virtual Memory
Demand Paging

 Bring a page into memory only when it is needed

 Less I/O needed
 Less memory needed
 Faster response
 More users

 Page is needed  reference to it

 invalid reference  abort
 not-in-memory  bring to memory
 Lazy swapper – never swaps a page into memory
unless page will be needed
 Swapper that deals with pages is a pager
Transfer of a Paged Memory to Contiguous Disk
Space
Valid-Invalid Bit
 With each page table entry a valid–invalid bit is associated
(v  in-memory, i  not-in-memory)
 Initially valid–invalid bit is set to i on all entries
 Example of a page table snapshot:
Frame # valid-invalid bit
v
v
v
v
i
….

i
i
page table

 During address translation, if valid–invalid bit in page table entry

is I  page fault
Page Table When Some Pages Are Not in Main
Memory
Page Fault
 If there is a reference to a page, first reference to that
page will trap to operating system:
page fault
1. Operating system looks at another table to decide:
 Invalid reference  abort
 Just not in memory
2. Get empty frame
3. Swap page into frame
4. Reset tables
5. Set validation bit = v
6. Restart the instruction that caused the page fault
Page Fault (Cont.)
 Restart instruction
 block move

 auto increment/decrement location

Steps in Handling a Page
Fault
Performance of Demand
Paging
 Page Fault Rate 0  p  1.0
 if p = 0 no page faults
 if p = 1, every reference is a fault

 Effective Access Time (EAT)

EAT = (1 – p) x memory access
+ p (page fault overhead
+ swap page out
+ swap page in
+ restart overhead
)
Demand Paging Example

 Memory access time = 200 nanoseconds

 Average page-fault service time = 8 milliseconds

 EAT = (1 – p) x 200 + p (8 milliseconds)

= (1 – p x 200 + p x 8,000,000
= 200 + p x 7,999,800

 If one access out of 1,000 causes a page fault, then

EAT = 8.2 microseconds.
This is a slowdown by a factor of 40!!
Process Creation

 Virtual memory allows other benefits during

process creation:

- Copy-on-Write

- Memory-Mapped Files (later)

Copy-on-Write

 Copy-on-Write (COW) allows both parent and child

processes to initially share the same pages in
memory

If either process modifies a shared page, only

then is the page copied

 COW allows more efficient process creation as

only modified pages are copied

 Free pages are allocated from a pool of zeroed-

out pages
Before Process 1 Modifies
Page C
After Process 1 Modifies
Page C
What happens if there is no free
frame?
 Page replacement – find some page in memory, but
not really in use, swap it out
 algorithm
 performance – want an algorithm which will result in
minimum number of page faults
 Same page may be brought into memory several
times
Page Replacement

 Prevent over-allocation of memory by modifying

page-fault service routine to include page
replacement

 Use modify (dirty) bit to reduce overhead of

page transfers – only modified pages are written
to disk

 Page replacement completes separation between

logical memory and physical memory – large
virtual memory can be provided on a smaller
physical memory
Need For Page
Replacement
Basic Page Replacement
1. Find the location of the desired page on disk

2. Find a free frame:

- If there is a free frame, use it
- If there is no free frame, use a page
replacement algorithm to select a victim
frame

3. Bring the desired page into the (newly) free

frame; update the page and frame tables

4. Restart the process

Page Replacement
Page Replacement
Algorithms
Want lowest page-fault rate

 Evaluate algorithm by running it on a particular

string of memory references (reference string) and
computing the number of page faults on that string

 In all our examples, the reference string is

1, 2, 3, 4, 1, 2, 5, 1, 2, 3, 4, 5
Graph of Page Faults Versus The
Number of Frames
First-In-First-Out (FIFO)
Reference string: 1, 2, 3, 4, 1, 2, 5, 1, 2, 3, 4, 5
Algorithm
3 frames (3 pages can be in memory at a time per process)

1 1 4 5
2 2 1 3 9 page faults
3 3 2 4

 4 frames
1 1 5 4
2 2 1 5 10 page faults
3 3 2

4 4 3

 Belady’s Anomaly: more frames  more page faults

FIFO Page Replacement
FIFO Illustrating Belady’s
Anomaly
Optimal Algorithm

 Replace page that will not be used for longest period

of time
 4 frames example 1 4
1, 2, 3,24, 1, 2, 5, 1, 2, 3, 4, faults
6 page 5

4 5

 How do you know this?

 Used for measuring how well your algorithm performs
Optimal Page
Replacement
Least Recently Used (LRU)
Algorithm

Reference string: 1, 2, 3, 4, 1, 2, 5, 1, 2, 3, 4, 5

1 1 1 1 5
2 2 2 2 2
3 5 5 4 4
4 4 3 3 3

 Counter implementation
 Every page entry has a counter; every time page is referenced
through this entry, copy the clock into the counter
 When a page needs to be changed, look at the counters to
determine which are to change
LRU Page Replacement
LRU Algorithm (Cont.)

 Stack implementation – keep a stack of page

numbers in a double link form:
 Page referenced:
 move it to the top
 requires 6 pointers to be changed
 No search for replacement
Use Of A Stack to Record The Most Recent Page
References
LRU Approximation
Algorithms

Reference bit

With each page associate a bit, initially = 0
 When page is referenced bit set to 1
 Replace the one which is 0 (if one exists)
 We do not know the order, however
 Second chance
 Need reference bit
 Clock replacement
 If page to be replaced (in clock order) has reference bit = 1
then:
 set reference bit 0
 leave page in memory
 replace next page (in clock order), subject to same rules
Second-Chance (clock) Page-Replacement Algorithm
Counting Algorithms
 Keep a counter of the number of references that have
been made to each page

 LFU Algorithm: replaces page with smallest count

 MFU Algorithm: based on the argument that the

page with the smallest count was probably just
brought in and has yet to be used
Allocation of Frames
 Each process needs minimum number of pages
 Example: IBM 370 – 6 pages to handle SS MOVE
instruction:
 instruction is 6 bytes, might span 2 pages
 2 pages to handle from
 2 pages to handle to
 Two major allocation schemes
 fixed allocation
 priority allocation
Fixed Allocation
 Equal allocation – For example, if there are 100 frames and
5 processes, give each process 20 frames.
 Proportional allocation – Allocate according to the size of
process
si size of process pi
S   si
m total number of frames
s
ai allocation for pi  i m
S
m 64
si 10
s2 127
10
a1  64 5
137
127
a2  64 59
137
Priority Allocation
 Use a proportional allocation scheme using priorities
rather than size

 If process Pi generates a page fault,

 select for replacement one of its frames
 select for replacement a frame from a process with lower
priority number
Global vs. Local Allocation
 Global replacement – process selects a
replacement frame from the set of all frames; one
process can take a frame from another
 Local replacement – each process selects from
only its own set of allocated frames
Thrashing
 If a process does not have “enough” pages, the page-fault
rate is very high. This leads to:
 low CPU utilization
 operating system thinks that it needs to increase the degree
of multiprogramming
 another process added to the system

 Thrashing  a process is busy swapping pages in and out

Thrashing (Cont.)
Demand Paging and Thrashing

 Why does demand paging work?

Locality model
 Process migrates from one locality to another
 Localities may overlap

 Why does thrashing occur?

 size of locality > total memory size
Locality In A Memory-Reference
Pattern
Working-Set Model
   working-set window  a fixed number of page
references
Example: 10,000 instruction
 WSSi (working set of Process Pi) =
total number of pages referenced in the most recent 
(varies in time)
 if  too small will not encompass entire locality
 if  too large will encompass several localities
 if  =   will encompass entire program
 D =  WSSi  total demand frames
 if D > m  Thrashing
 Policy if D > m, then suspend one of the processes
Working-set model
Keeping Track of the
Working Set
 Approximate with interval timer + a reference bit
 Example:  = 10,000
 Timer interrupts after every 5000 time units
 Keep in memory 2 bits for each page
 Whenever a timer interrupts copy and sets the
values of all reference bits to 0
 If one of the bits in memory = 1  page in working
set
 Why is this not completely accurate?
 Improvement = 10 bits and interrupt every 1000
time units
Page-Fault Frequency
Scheme
Establish “acceptable” page-fault rate
 If actual rate too low, process loses frame
 If actual rate too high, process gains frame
Working Sets and Page Fault Rates
Memory-Mapped Files

 Memory-mapped file I/O allows file I/O to be treated

as routine memory access by mapping a disk block
to a page in memory

 A file is initially read using demand paging. A page-

sized portion of the file is read from the file system
into a physical page. Subsequent reads/writes
to/from the file are treated as ordinary memory
accesses.

 Simplifies file access by treating file I/O through

memory rather than read() write() system calls

 Also allows several processes to map the same file

allowing the pages in memory to be shared
Memory Mapped Files
Memory-Mapped Shared Memory
in Windows
Allocating Kernel Memory

 Treated differently from user memory

 Often allocated from a free-memory pool
 Kernel requests memory for structures of varying
sizes
 Some kernel memory needs to be contiguous
Buddy System

 Allocates memory from fixed-size segment

consisting of physically-contiguous pages
 Memory allocated using power-of-2 allocator
 Satisfies requests in units sized as power of 2
 Request rounded up to next highest power of 2
 When smaller allocation needed than is available,
current chunk split into two buddies of next-lower
power of 2
 Continue until appropriate sized chunk available
Buddy System Allocator
Slab Allocator

 Alternate strategy
 Slab is one or more physically contiguous pages
 Cache consists of one or more slabs
 Single cache for each unique kernel data structure
 Each cache filled with objects – instantiations of the
data structure
 When cache created, filled with objects marked as
free
 When structures stored, objects marked as used
 If slab is full of used objects, next object allocated
from empty slab
 If no empty slabs, new slab allocated
 Benefits include no fragmentation, fast memory
request satisfaction
Slab Allocation
Other Issues -- Prepaging
 Prepaging
 To reduce the large number of page faults that occurs at
process startup
 Prepage all or some of the pages a process will need, before
they are referenced
 But if prepaged pages are unused, I/O and memory was wasted
 Assume s pages are prepaged and α of the pages is used
 Is cost of s * α save pages faults > or < than the cost of prepaging
s * (1- α) unnecessary pages?
 α near zero  prepaging loses
Other Issues – Page Size
 Page size selection must take into consideration:
 fragmentation
 table size
 I/O overhead
 locality
Other Issues – TLB Reach
 TLB Reach - The amount of memory accessible from
the TLB
 TLB Reach = (TLB Size) X (Page Size)
 Ideally, the working set of each process is stored in
the TLB
 Otherwise there is a high degree of page faults
 Increase the Page Size
 This may lead to an increase in fragmentation as not all
applications require a large page size
 Provide Multiple Page Sizes
 This allows applications that require larger page sizes the
opportunity to use them without an increase in
fragmentation
Other Issues – Program
Structure
 Program structure
Int[128,128] data;
 Each row is stored in one page
 Program 1
for (j = 0; j <128; j++)
for (i = 0; i < 128; i++)
data[i,j] = 0;

128 x 128 = 16,384 page faults

 Program 2
for (i = 0; i < 128; i++)
for (j = 0; j < 128; j++)
data[i,j] = 0;

128 page faults

Other Issues – I/O interlock
 I/O Interlock – Pages must sometimes be locked
into memory

 Consider I/O - Pages that are used for copying a

file from a device must be locked from being
selected for eviction by a page replacement
algorithm
Reason Why Frames Used For I/O Must Be In
Memory
Operating System
Examples
 Windows XP

 Solaris
Windows XP

 Uses demand paging with clustering. Clustering

brings in pages surrounding the faulting page
 Processes are assigned working set minimum and
working set maximum
 Working set minimum is the minimum number of
pages the process is guaranteed to have in memory
 A process may be assigned as many pages up to its
working set maximum
 When the amount of free memory in the system falls
below a threshold, automatic working set trimming
is performed to restore the amount of free memory
 Working set trimming removes pages from processes
that have pages in excess of their working set
minimum
Solaris

 Maintains a list of free pages to assign faulting

processes
 Lotsfree – threshold parameter (amount of free
memory) to begin paging
 Desfree – threshold parameter to increasing paging
 Minfree – threshold parameter to being swapping
 Paging is performed by pageout process
 Pageout scans pages using modified clock algorithm
 Scanrate is the rate at which pages are scanned.
This ranges from slowscan to fastscan
 Pageout is called more frequently depending upon
the amount of free memory available
Solaris 2 Page Scanner
End of Chapter 9