Originally posted by: SystemAdmin
This status bother me for while and do not know how to tackle it. Here is a snapshot of my AIX 5.3L p550 (4g memory ,2 CPU)
Topas Monitor for host: baanserver EVENTS/QUEUES FILE/TTY
Fri May 25 07:58:51 2007 Interval: 2 Cswitch 412 Readch 178.3K
Syscall 2090 Writech 47877
Kernel 0.9 |# | Reads 161 Rawin 0
User 2.4 |# | Writes 149 Ttyout 426
Wait 0.3 |# | Forks 0 Igets 0
Idle 96.4 |############################| Execs 0 Namei 42
Runqueue 0.0 Dirblk 0
Network KBPS I-Pack O-Pack KB-In KB-Out Waitqueue 0.0
en0 8.5 9.0 11.5 0.4 8.1
lo0 0.9 3.0 3.0 0.5 0.5 PAGING MEMORY
Faults 0 Real,MB 3792
Disk Busy% KBPS TPS KB-Read KB-Writ Steals 0 % Comp 55.6
hdisk0 1.5 2.0 0.5 0.0 2.0 PgspIn 0 % Noncomp 44.7
hdisk2 0.0 0.0 0.0 0.0 0.0 PgspOut 0 % Client 46.0
hdisk3 0.0 0.0 0.0 0.0 0.0 PageIn 0
hdisk1 0.0 0.0 0.0 0.0 0.0 PageOut 0 PAGING SPACE
cd0 0.0 0.0 0.0 0.0 0.0 Sios 0 Size,MB 6272
% Used 44.3
Name PID CPU% PgSp Owner NFS (calls/sec) % Free 55.6
-
lsps -a
Page Space Physical Volume Volume Group Size %Used Active Auto Type
hd6 hdisk0 rootvg 6272MB 44 yes yes lv
-
svmon -G
size inuse free pin virtual
memory 970752 963774 6978 247720 1061833
pg space 1605632 703340
work pers clnt
pin 247720 0 0
in use 526068 0 437706
I noticed that the %use nad clnt (%client) value are high especially in the morning before majority of employees began to use the system (a BAAN and Oracle 10gR2). These values will gradually drop during the business hours, 10 AM ~ 4 PM, when more user are on. Do I misunderstand the meanings of %use and %client or they did indicate something needed to tune? How do I start?