Modeling Page Replacement Algorithms
I.S.L Sarwani, Asst prof, IT, ANITS
Modeling Page Replacement Algorithms Belady's Anomaly
FIFO with 3 page frames FIFO with 4 page frames P's show which page references show page faults
Stack Algorithms
State of memory array, M, after each item in reference string is processed
Contd..
Paging system can be categorized by 3 items: 1. The reference string of executing process 2. the page replacement algorithm 3. the number of page frames available in memory i.e. m M- internal array keeps track of state of memory Top part- all the page entries currently in memory Bottom part- pages referenced once, swapped out and are not in memory
The Distance String
A page reference will be denoted by the distance from top of stack where reference page was located. Pages that have not yet been reference distance infinity. Distance string depends on reference string and paging algorithm.
Predicting page fault rates
Computation of page fault rate from distance string
the C vector the F vector
Contd..
C number of occurrences of i or infinity in distance string F number of page faults
Design issues for paging systems
Local versus Global Allocation Policies (1)
Original configuration Local page replacement Global page replacement
Local when a process gets page fault , the page replacement algorithm try to find the LRU page considering pages that are currently allocated to only that process. Global if the page with lowest age value is removed without regard to whose page it is.
Contd..
Local allocates fixed memory to process. If working set grows thrashing will result. So wasting memory. Global dynamical allocation among runnable processes. Here OD continually decides how many page frames to assign to each process. Page Fault Frequency (PFF) algorithm: Tells when to increase or decrease process page allocation but not which page to remove
Load Control
Despite good designs, system may still thrash When PFF algorithm indicates
some processes need more memory but no processes need less
Solution : Reduce number of processes competing for memory
swap one or more to disk, divide up pages they held reconsider degree of multiprogramming
Page Size (1)
Small page size Advantages
less internal fragmentation better fit for various data structures, code sections less unused program in memory
Disadvantages
programs need many pages, larger page tables
Separate Instruction and Data Spaces
One address space Separate I and D spaces
Shared Pages
Two processes sharing same program sharing its page table
Contd..
If I space and D space are supported 2 or more processes use the same page table for their I space but different D space. Processes share same code problem occurs Copy-on-write: Those pages that are never written need not be copied.
Cleaning Policy
Need for a background process, paging daemon
periodically inspects state of memory
When too few frames are free
selects pages to evict using a replacement algorithm
It can use same circular list (clock) front hand controlled by paging daemon Back hand page replacement
Segmentation
Segmentation (1)
One-dimensional address space with growing tables One table may bump into another
Segmentation (2)
Allows each table to grow or shrink, independently
Segmentation (3)
Comparison of paging and segmentation
Implementation of Pure Segmentation
(a)-(d) Development of checkerboarding (e) Removal of the checkerboarding by compaction
