Information Technology Reference
In-Depth Information
Figure 10. The memory performance of bit-r and gcc in their concurrent execution managed with swap
token.
global LRU to reflect memory access patterns of
processes. The measurements show that the ex-
ecution times of both programs are significantly
reduced by the swap token compared with the
times in the original Linux LRU. The slowdown
of
bit-r
is 2.08 (a reduction of 23%), and is 2.25
for
gcc
(a reduction of 38%). The page fault
reductions for
bit-r
and
gcc
are 20% and 82%,
respectively.
Figure 11 presents the memory performance
measured by MAD and RSS of concurrently
running programs
gcc
and
vortex
when the swap
token is introduced. At the execution time of 397
th
second, both processes started page faults due to
memory shortage. The token was taken by
gcc
after this time.
Figure 11 shows that
gcc
quickly built up its
working set, reflected by keeping the first RSS
Figure 11. The memory performance of gcc and vortex in their concurrent execution managed with
swap token.
Search WWH ::
Custom Search