[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Benchmark] some 2.5 scheduler backporting to ck4
Hi again,
ck4 is ck4 + rmap15d + (rmap15e incrememtal - elevator.h stuff)
ck4-np1 is the above with the backports patch in the previous email
no_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 5 61 93.4 0.0 0.0 1.00
2.4.20-ck4-np1 3 62 93.5 0.0 0.0 1.00
cacherun:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 5 58 98.3 0.0 0.0 0.95
2.4.20-ck4-np1 3 58 100.0 0.0 0.0 0.94
process_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 5 91 61.5 95.4 9.9 1.49
2.4.20-ck4-np1 3 199 29.1 409.0 70.4 3.21
ctar_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 5 76 77.6 3.0 5.3 1.25
2.4.20-ck4-np1 3 84 71.4 6.3 10.7 1.35
xtar_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 5 80 72.5 1.4 8.8 1.31
2.4.20-ck4-np1 3 85 69.4 2.3 10.6 1.37
io_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 5 87 66.7 13.4 11.4 1.43
2.4.20-ck4-np1 3 107 55.1 15.7 11.0 1.73
io_other:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 5 85 68.2 15.0 14.1 1.39
2.4.20-ck4-np1 3 105 56.2 16.4 13.2 1.69
read_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 5 90 66.7 4.3 6.7 1.48
2.4.20-ck4-np1 3 96 64.6 6.5 9.4 1.55
list_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 5 76 75.0 0.0 7.9 1.25
2.4.20-ck4-np1 3 82 72.0 0.0 13.4 1.32
mem_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 5 95 63.2 92.2 4.2 1.56
2.4.20-ck4-np1 3 111 55.0 95.7 3.6 1.79
dbench_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.4.20-ck4 4 172 29.1 3.8 53.3 2.82
2.4.20-ck4-np1 3 234 17.5 6.3 71.5 3.77
Some weirdness going on in dbench_load for one of the results.
ck4 is short one run because I ran out if disk space and didn't want to
run it again
==> dbench_load.log <==
2.4.20-ck4-np1 55 3 352 319900 259905 dbench_load 32 224 352 11468 58059 1000
2.4.20-ck4-np1 55 3 322 319900 259897 dbench_load 28 204 322 10324 52255 900
2.4.20-ck4-np1 9 0 28 56217 42942 dbench_load 0 1 11 197 840 0
2.4.20-ck4 55 3 210 319899 259895 dbench_load 13 94 210 4722 23803 500
2.4.20-ck4 55 3 204 319899 259895 dbench_load 16 104 204 5739 29039 500
2.4.20-ck4 56 3 230 319899 259895 dbench_load 15 112 230 5740 29039 500
2.4.20-ck4 27 1 45 158672 118119 dbench_load 0 0 19 129 484 0
ck4 also stalled once during proces_load, STOPping and CONTinuing it
worked.
==> process_load.log <==
2.4.20-ck4-np1 55 3 203 319893 259905 process_load 9 134 203 32 15 42300
2.4.20-ck4-np1 55 3 196 319900 259897 process_load 9 128 196 33 15 40400
2.4.20-ck4-np1 55 3 200 319900 259895 process_load 9 132 200 34 15 40000
2.4.20-ck4 54 3 91 319899 259895 process_load 2 30 1313 33 15 9600
2.4.20-ck4 54 2 91 319898 259895 process_load 2 31 91 34 15 9800
2.4.20-ck4 54 2 93 319898 259895 process_load 2 32 93 33 15 9400
2.4.20-ck4 54 2 91 319897 259895 process_load 2 31 91 34 15 9500
2.4.20-ck4 54 2 91 319899 259895 process_load 2 31 91 35 15 9400
All the other runs finished with consistent times
--
Eric Wong
--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive: http://mail.nl.linux.org/kernelnewbies/
FAQ: http://kernelnewbies.org/faq/