[osdl-aim-7] 2.6.0-test5-mm2

From: Cliff White
Date: Mon Sep 15 2003 - 16:33:44 EST



Here are results from OSDL's
Scalable Test Platform.

Performance falls off compared to -test5 on the
8-cpu machines.

We are also experiencing some warnings on some tests:

Badness in as_dispatch_request at drivers/block/as-iosched.c:1241

And, we have three machines with processes hanging in 'D' state.
More data as we get it.
cliffw

-----------------------------
Database workload

stp1 CPU machine

STP id PLM# Kernel Name Workfile MaxJPM MaxU Change Host
Options
Newest Kernel - Baseline for % change
279938 2144 2.6.0-test5-mm2 new_dbase 999.53 17 0.00
stp1-001 profile=2
279569 2112 2.6.0-test5-mm1-fix11.0 new_dbase 990.94 17 -0.86
stp1-003 profile=2
279455 2110 linux-2.6.0-test5 new_dbase 992.06 17 -0.75
stp1-003 profile=2
stp 2-cpu
279957 2144 2.6.0-test5-mm2 new_dbase 1306.85 22 0.00
stp2-002 profile=2
279879 2142 2.6.0-test5-O1int20.1-2 new_dbase 1325.89 22 1.46
stp2-003 profile=2
279474 2110 linux-2.6.0-test5 new_dbase 1337.11 22 2.32
stp2-000 profile=2
stp 8-cpu
279931 2144 2.6.0-test5-mm2 new_dbase 8273.17 136 0.00
stp8-003 profile=2 elevator=cfq
279929 2144 2.6.0-test5-mm2 new_dbase 8309.33 136 0.44
stp8-003 profile=2 elevator=deadline
279927 2144 2.6.0-test5-mm2 new_dbase 8547.23 136 3.31
stp8-000 profile=2
279448 2110 linux-2.6.0-test5 new_dbase 8812.21 144 6.52
stp8-000 profile=2 elevator=cfq
279446 2110 linux-2.6.0-test5 new_dbase 8950.07 144 8.18
stp8-002 profile=2 elevator=deadline
279444 2110 linux-2.6.0-test5 new_dbase 8785.24 144 6.19
stp8-002 profile=2


---------------
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)

cliffw



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/