Browse Prior Art Database

Page Storage Algorithm

IP.com Disclosure Number: IPCOM000117005D
Original Publication Date: 1995-Dec-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 2 page(s) / 25K

Publishing Venue

IBM

Related People

Brew, GE: AUTHOR [+2]

Abstract

Disclosed is a method in which a demand paging operating system attempts to compress a memory page before swapping it out to a disk. If the page can be compressed to a predetermined size or less, the page is compressed and stored in a reserved area of resident memory. If the page cannot be compressed to this predetermined size, the page is swapped out to disk to make space available for other pages.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 100% of the total text.

Page Storage Algorithm

      Disclosed is a method in which a demand paging operating system
attempts to compress a memory page before swapping it out to a disk.
If the page can be compressed to a predetermined size or less, the
page is compressed and stored in a reserved area of resident memory.
If the page cannot be compressed to this predetermined size, the page
is swapped out to disk to make space available for other pages.

      This method avoids unnecessary instances of writing pages to
disk, reducing the overhead otherwise incurred by disk writes and
reads to the swapfile.  However, the compressed size of a page must
be significantly smaller than its uncompressed size for savings to
occur; otherwise, the reserved memory area needed for this process is
large enough to cause increased swapping, offsetting the savings
accrued through the use of this method.  For example, significant
savings have been demonstrated with a 64:1 compression ratio.