The performance of memory-intensive applications tends to be poor due to th
e high overhead added by the swapping mechanism. The same problem may be fo
und in highly-loaded multi-programming systems where all running applicatio
ns have to use the swap space in order to be able to execute at the same ti
me. In this paper, we present a solution to these problems. The idea consis
ts of compressing the swapped pages and keeping them in a swap cache whenev
er possible.
The idea of a compressed swap cache was proposed a few years ago, but it di
d not achieve the expected results due to hardware limitations. As processo
rs are much faster and the performance gap between processors and disk is a
lso much bigger, me believe that now it is the right time to revisit this i
dea. In this paper, we present a new implementation of the compressed cache
, plus some enhancements that significantly improve the performance of the
swapping mechanism. Copyright (C) 2000 John Wiley & Sons, Ltd.