All results: http://developer.osdl.org/cliffw/reaim/index.html
I am now running two workloads, new_dbase and compute:
The compute workload has much less IO, more system time.
On the 1 and 2 cpu machines, very little delta, test4 a little better:
1-CPU
STP id PLM# Kernel Name Workfile MaxJPM MaxUser Host Change
278204 2070 2.6.0-test4-mm1 new_dbase 988.42 17 stp1-001 0.00
278000 2067 linux-2.6.0-test4 new_dbase 990.15 17 stp1-003 0.18
278205 2070 2.6.0-test4-mm1 compute 1011.76 17 stp1-003 0.00
278001 2067 linux-2.6.0-test4 compute 1019.36 17 stp1-000 0.75
2-CPU
278215 2070 2.6.0-test4-mm1 new_dbase 1325.86 22 stp2-000 0.00
278011 2067 linux-2.6.0-test4 new_dbase 1331.56 24 stp2-000 0.43
278216 2070 2.6.0-test4-mm1 compute 1511.41 26 stp2-001 0.00
278012 2067 linux-2.6.0-test4 compute 1535.87 26 stp2-001 1.62
Running on the 8-cpu machines, delta is larger:
STP id PLM# Kernel Name Workfile MaxJPM MaxUser Host Change
278184 2070 2.6.0-test4-mm1 new_dbase 8203.60 136 stp8-001 0.00
277980 2067 linux-2.6.0-test4 new_dbase 8483.27 144 stp8-002 3.41
278186 2070 2.6.0-test4-mm1 compute 9601.96 160 stp8-003 0.00
277982 2067 linux-2.6.0-test4 compute 9432.60 160 stp8-000 -1.76
---------------
Detail on any run:
http://khack.osdl.org/stp/<STP id>
Hardware details:
http://khack.osdl.org/stp/<STP id>/environment/machine_info
More results:
http://developer.osdl.org/cliffw/reaim/index.html
---------------
Code location:
bk://developer.osdl.org/osdl-aim-7
tarball:
http://sourceforge.net/projects/re-aim-7
Run parameters:
./reaim -s$CPU_COUNT -x -t -i$CPU_COUNT -f workfile.new_dbase -r3 -b
-l./stp.config
./reaim -s$CPU_COUNT -q -t -i$CPU_COUNT -f workfile.new_dbase -r3 -b
-l./stp.config
(3 runs each, average of all 6 reported runs repeated with workfile.compute)
cliffw