Originally posted by: aks1
Hi Guys
I am running performance test suite on my application with Weblogic11(10.3.6) and Oracle 11g DB(11.2.0.4) on AIX7.1. While CPU and memory utilization on app server and DB srever seems normal, response time is not coming down, as per DBA there no significant scope for query tuning.
We have captured Nmon logs and per NMon graphs(vai nMon Analyser), i saw some spikes after 12-13 mins on DB sever, thse spikes are visible in memory,CPU and Disk busy stats. Also on the paging sheet, there's significant paging out but no scanning.
Report Time
|
Base Vol
|
User
|
Server
|
VCPUS
|
RAM(GB)
|
Avg %
|
Max %
|
Avg %
|
Max %
|
Max DB Seesion
|
DB
Updates
|
Txn>5Sec
|
11:45 to 13:15
|
10L
|
500
|
App Server
|
12
|
24
|
16
|
38
|
|
|
338
|
96.094
|
35.26%
|
DB Server
|
12
|
24
|
39.5
|
86
|
|
|
NMon Analyser Graphs(DB Server):
Here are some of the vmo parameters from DB Server
bash-3.2# oslevel -r
7100-03
bash-3.2# lsps -s
Total Paging Space Percent Used
16384MB 0%
bash-3.2# swap -l
device maj,min total free
/dev/hd6 10, 2 16384MB 16315MB
bash-3.2# lsps -a
Page Space Physical Volume Volume Group Size %Used Active Auto Type Chksum
hd6 hdisk0 rootvg 16384MB 0 yes yes lv 0
bash-3.2# uptime
11:40AM up 255 days, 16:46, 5 users, load average: 1.29, 1.28, 1.27
bash-3.2# vmo -o defps
defps = 1
vmo parameters
minfree = 960
maxfree = 1088
minperm = 180899
maxperm = 5426999
maxpin = 5687744
minperm% = 3
numperm%= 56.6
maxpin% = 90
maxperm% = 90 (vmstat -v)
maxclient%=90 (vmstat -v)
numclient%= 56.6 (vmstat -v)
low_ps_handling = 1
j2_maxPageReadAhead = 128(ioo -a -F | grep head)
j2_minPageReadAhead = 2
maxpgahead = 8
minpgahead = 2
bash-3.2# vmstat -v shows 2 memory pools
Some key entries from MEM sheet of nMon
Paging AIX4-11-13
|
faults
|
pgin
|
pgout
|
pgsin
|
pgsout
|
reclaims
|
scans
|
cycles
|
fsin
|
fsout
|
sr/fr
|
11:50:59
|
6944
|
0.6
|
1321.8
|
0
|
0
|
0
|
0
|
0
|
0.6
|
1321.8
|
0.0
|
11:51:04
|
3450.1
|
2.8
|
3935.1
|
0
|
0
|
0
|
0
|
0
|
2.8
|
3935.1
|
0.0
|
11:52:49
|
3669.4
|
4.4
|
8051.5
|
0
|
0
|
0
|
0
|
0
|
4.4
|
8051.5
|
0.0
|
12:07:59
|
1778.4
|
11.8
|
1726.7
|
0
|
0
|
0
|
0
|
0
|
11.8
|
1726.7
|
0.0
|
12:08:04
|
859
|
6.4
|
6592.6
|
0
|
0
|
0
|
0
|
0
|
6.4
|
6592.6
|
0.0
|
12:21:14
|
2280.8
|
1
|
8084.4
|
0
|
0
|
0
|
0
|
0
|
1.0
|
8084.4
|
0.0
|
12:33:14
|
1523.7
|
0.8
|
8382
|
0
|
0
|
0
|
0
|
0
|
0.8
|
8382.0
|
0.0
|
12:47:04
|
1298.4
|
0.2
|
8236.5
|
0
|
0
|
0
|
0
|
0
|
0.2
|
8236.5
|
0.0
|
13:00:10
|
1038.6
|
0
|
9044
|
0
|
0
|
0
|
0
|
0
|
0.0
|
9044.0
|
0.0
|
13:12:40
|
3460.1
|
4.4
|
8349.4
|
0
|
0
|
0
|
0
|
0
|
4.4
|
8349.4
|
0.0
|
SGA memory details of the database are as follow:
SGA - 2GB
PGA - 0
Memory_Target - 3GB
Can some one tell me whats wrong going on the DB Srever?
Thanks in advance