page replacement algorithm powerpoint presentation

CMSC 421 Section 0202
Storage Management
Chapter 10: Virtual Memory

Silberschatz, Galvin and Gagne 2002
10.2
Operating System Concepts
Contents
 Background
 Demand Paging
 Process Creation
 Page Replacement
 Allocation of Frames
 Thrashing
 Operating System Examples

10.3
Background
 Virtual memory
 enables the separation of user logical memory from physical
memory.
 moreover
 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

10.4
Virtual Memory That is Larger Than Physical Memory

10.5
Demand Paging
 Pager brings a page into memory only when it is needed.
 Less I/O needed
 Less memory needed
 Faster response (since all pages of a process are not
swapped in)
 More users
 Page is needed  there is a (memory) reference to it
 invalid reference  abort
 not-in-memory  bring to memory

10.6
Transfer of a Paged Memory to Contiguous Disk Space

10.7
Valid-Invalid Bit
 With each page table entry a valid–invalid bit is associated
(1  in-memory, 0  not-in-memory)
 Initially valid–invalid but is set to 0 on all entries.
 Example of a page table snapshot.
 During address translation, if valid–invalid bit in page table entry is 0 
page fault.
1
1
1
1
0
0
0

Frame # valid-invalid bit
page table

10.8
Page Table When Some Pages Are Not in Main Memory

10.9
Page Fault
 If there is ever a reference to a page, first reference will trap to
OS  page fault
 OS looks at another table to decide:
 Invalid reference  abort.
 Just not in memory.
 Get empty frame.
 Swap page into frame.
 Reset tables, valid bit = 1.
 Restart instruction
 Problems
 block move
 auto increment/decrement location

10.10
Steps in Handling a Page Fault

10.11
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.
 In general to implement demand paging we need
 A Frame-allocation algorithm
 A page-replacement algorithm

10.12
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])

10.13
Demand Paging Example
 Memory access time = 1 microsecond
 50% of the time the page that is being replaced has been
modified and therefore needs to be swapped out.
 Swap Page Time = 10 msec = 10,000 microseconds
EAT = (1 – p) x 1 + p (10000+0.5X10000)
1 + 15000p (in msec)

10.14
Process Creation
 Virtual memory allows other benefits during process
creation:
- Copy-on-Write
- Memory-Mapped Files

10.15
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.
 Zeroed-out pages are pages whose contents are erased
before they are allocated

10.16
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.

10.17
Memory Mapped Files

10.18
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.

10.19
Need For Page Replacement

10.20
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. Read the desired page into the (newly) free frame.
Update the page and frame tables.
4. Restart the process.

10.21
Page Replacement

10.22
Page Replacement Algorithms
 Want lowest page-fault rate.
 Evaluate algorithm by
 running it on a particular string of memory references
(reference string) and then
 computing the number of page faults on that string.

10.23
Graph of Page Faults Versus The Number of Frames

10.24
First-In-First-Out (FIFO) Algorithm
 Reference string: 1, 2, 3, 4, 1, 2, 5, 1, 2, 3, 4, 5
 3 frames (3 pages can be in memory at a time per process)
 4 frames
 FIFO Replacement – Belady’s Anomaly
 more frames  less page faults
1
2
3
1
2
3
4
1
2
5
3
4
9 page faults
1
2
3
1
2
3
5
1
2
4
5 10 page faults
4
4 3

10.25
FIFO Page Replacement

10.26
FIFO Illustrating Belady’s Anamoly

10.27
Optimal Algorithm
 Replace page that will not be used for longest period of time.
 4 frames example
1, 2, 3, 4, 1, 2, 5, 1, 2, 3, 4, 5
 How do you know this?
 Used for measuring how well your algorithm performs.
1
2
3
4
6 page faults
4 5

10.28
Optimal Page Replacement

10.29
Least Recently Used (LRU) Algorithm
 Reference string: 1, 2, 3, 4, 1, 2, 5, 1, 2, 3, 4, 5
 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 (in increasing clock order)
 Requires linear search of all the frames
1
2
3
5
4
4 3
5

10.30
LRU Page Replacement

10.31
LRU Algorithm (Cont.)
 Stack implementation
 keep a stack of page numbers in a double link form:
 Whenever a page is referenced
 move it to the top
 requires 6 pointers to be changed
 No search for replacement

10.32
Use Of A Stack to Record The Most Recent Page References

10.33
Stack Page-Replacement Algorithms
 It is an algorithm that the set of pages in memory for n
frames is always a subset of the set of pages in memory
for n+1 frames
 They never exhibit Belady’s anomaly
 Example Stack algorithms
 Optimal
 LRU

10.34
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.
 Scheme can be extended to multiple reference bits (as a page reference)
 Set high-order bit to 1 at each reference; right shift periodically, replace
page with smallest value)
 Second chance (Clock algorithm)
 Need reference bit and Clock (page arrival time)
 If page to be replaced (in clock order) has reference bit = 1. then:
 set reference bit 0 and reset its clock value.
 leave page in memory.
 replace next page (in clock order), subject to same rules.
 Can also be implemented using a circular queue of pages
 Can be extended to also consider a “dirty” bit

10.35
Second-Chance (clock) Page-Replacement Algorithm

10.36
Counting Algorithms
 Keeps a counter of the number of references that have
been made to each page within a certain time window
 LFU Algorithm
 replaces page with smallest count.
 MFU Algorithm
 Replace page with largest count
 It is based on the argument that the page with the smallest
count was probably just brought in and has yet to be used.

10.37
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

10.38
Fixed Allocation
 Equal allocation – e.g., if 100 frames and 5
processes, give each 20 pages.
 Proportional allocation – Allocate according to
the size of process.
m
S
s
p
a
m
s
S
p
s
i
i
i
i
i
i







for
allocation
frames
of
number
total
process
of
size
59
64
137
127
5
64
137
10
127
10
64
2
1
2









a
a
s
s
m
i

10.39
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.

10.40
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.
 Global replacement offers better throughput
 Local replacement offers lower variance in response time

10.41
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.

10.42
Thrashing
 Why does paging work?
Locality model
 Process migrates from one locality to another.
 Localities may overlap.
 Why does thrashing occur?
 size of locality > total memory size

10.43
Locality In A Memory-Reference Pattern

10.44
Working-Set Model
   working-set window  a fixed number of page
references
Example: 10,000 instructions (or memory references)
 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.

10.45
Working-set model

10.46
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 set 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.

10.47
Page-Fault Frequency Scheme
 Direct control for thrashing
 Establish “acceptable” page-fault rates.
 If actual rate too low, process loses frame.
 If actual rate too high, process gains frame.
 If no free frames, reduce level of multiprogramming

10.48
Other Considerations
 Prepaging
 Load a set of pages that expect to be needed
 Page size selection
 Internal fragmentation
 Smaller pages lead to less fragmentation
 table size
 Smaller page size leads to larger page tables
 I/O overhead
 I/O with 1 large page is better than with 2 I/Os for pages of ½ size
 Larger page size leads to bringing in useless data
 Locality
 Smaller page sizes capture locality at higher resolution
 Page fault rate
 Smaller page sizes lead to higher page fault rates

10.49
Other Considerations (Cont.)
 TLB Reach
 The amount of memory accessible from the TLB.
 Related to hit ratio when translating virtual addresses
 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.

10.50
Increasing the Size of the TLB
 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.
 Manage TLB in software

10.51
 Program structure
 int A[][] = new int[1024][1024];
 Each row is stored in one page
 Program 1 for (j = 0; j < A.length; j++)
for (i = 0; i < A.length; i++)
A[i,j] = 0;
1024 x 1024 page faults
 Program 2 for (i = 0; i < A.length; i++)
for (j = 0; j < A.length; j++)
A[i,j] = 0;
1024 page faults

10.52
 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.

10.53
Reason Why Frames Used For I/O Must Be In Memory

10.54
Operating System Examples
 Windows NT
 Solaris 2

10.55
Windows NT
 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.

10.56
Solaris 2
 Maintains a list of free pages to assign faulting processes.
 Lotsfree – threshold parameter to begin paging.
 Paging is peformed by pageout process.
 Pageout scans pages using modified clock algorithm.
 Scanrate is the rate at which pages are scanned. This ranged
from slowscan to fastscan.
 Pageout is called more frequently depending upon the amount
of free memory available.

10.57
Solar Page Scanner

page replacement algorithm powerpoint presentation

More Related Content

Similar to page replacement algorithm powerpoint presentation (20)

Recently uploaded (20)

page replacement algorithm powerpoint presentation