high sys% in top
We have IDS 10.0 64bit running on HPUX 11.11, although the problem is not
really an IDS issue.
The DB is on a NAS which is configured raid 4, single fibre channel
controller [not my choice before you ask, I got TOLD what I needed!]. As far
as I can ascertain there are 6 drives for the system and the stripe size is
15mb [again out of my control]. Largely I am restricted to asking others to
take a look at configuration/performance issues related to the NAS but will
not get a straight answer when it comes to setup. I get told "not to worry
about it".
What concerns me is the high sysio figure. This happens with anything that
reads/writes to FS.
Basically I see it happening with programs that generate reports, perform
sorts or load data from ascii files.
Anyone got any thoughts on this? I expect to see sys climb when doing FS io,
but not this high. The amount of IO being performed at the time is not
excessive.
Instance is configured to use raw space and performance is more than
acceptable.
System: nh1jpscp Tue Nov 1 20:48:48
2005
Load averages: 2.61, 2.34, 2.10
227 processes: 205 sleeping, 22 running
Cpu states:
CPU LOAD USER NICE SYS IDLE BLOCK SWAIT INTR SSYS
0 2.50 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%
1 2.68 0.0% 0.0% 0.2% 99.8% 0.0% 0.0% 0.0% 0.0%
2 2.64 0.0% 0.0% 84.9% 15.1% 0.0% 0.0% 0.0% 0.0%
--- ---- ----- ----- ----- ----- ----- ----- ----- -----
avg 2.61 0.0% 0.0% 28.4% 71.6% 0.0% 0.0% 0.0% 0.0%
Memory: 1896464K (1664424K) real, 2131072K (1758820K) virtual, 423472K free
Pag
e# 1/23
CPU TTY PID USERNAME PRI NI SIZE RES STATE TIME %WCPU %CPU COMMAND
2 ? 7592 mutual 152 20 900K 868K sleep 1479:29 4.53 4.52
4gl_go.exe
0 ? 27076 fund 152 20 900K 868K sleep 501:13 4.50 4.49
4gl_go.exe
1 ? 25886 mutual 152 20 932K 900K sleep 1396:26 4.48 4.48
4gl_go.exe
1 ? 29643 hmhorne 152 20 2944K 1012K sleep 741:12 2.66 2.66
dbaccess
1 ? 20381 dldouil 152 20 2944K 1012K sleep 735:22 2.62 2.62
dbaccess
1 ? 27266 dldouil 152 20 2944K 1012K sleep 789:57 2.62 2.62
dbaccess
1 ? 20602 dldouil 152 20 2944K 1012K sleep 813:15 2.62 2.61
dbaccess
2 ? 18906 hmhorne 152 20 2944K 1012K sleep 730:42 2.61 2.61
dbaccess
1 ? 13696 dldouil 152 20 2944K 1012K sleep 793:36 2.56 2.55
dbaccess
iostat 1 5
device bps sps msps
c1t6d0 0 0.0 1.0
c8t0d1 0 0.0 1.0
c1t6d0 3 0.9 1.0
c8t0d1 126 31.1 1.0
c1t6d0 5 2.0 1.0
c1t6d0 0 2.0 1.0
c1t6d0 4 2.0 1.0
c8t0d1 13 4.0 1.0
sar -d 1 5
HP-UX nh1jpscp B.11.11 U 9000/800 11/01/05
21:03:21 device %busy avque r+w/s blks/s avwait avserv
21:03:22 c1t6d0 1.98 0.50 4 40 6.07 8.82
21:03:23 c1t6d0 1.00 0.50 1 4 6.41 11.32
21:03:24 c1t6d0 3.00 0.50 4 16 6.65 14.66
21:03:25 c1t6d0 1.00 0.50 1 4 6.64 7.22
21:03:26 c1t6d0 2.00 0.50 3 12 6.25 12.15
Average c1t6d0 1.80 0.50 3 15 6.36 11.45
Filesystem configuration. Vg00 are internal, mirrored. Vg01 is on NAS, same
disks IDS uses for db.
Filesystem kbytes used avail %used Mounted on
/dev/vg00/lvol3 409600 101032 306192 25% /
/dev/vg00/lvol1 298928 50752 218280 19% /stand
/dev/vg01/windward 25083904 29012 24663432 0% /windward
/dev/vg00/lvol8 4710400 510224 4167608 11% /var
/dev/vg00/lvol7 4194304 2156008 2022408 52% /usr
/dev/vg00/lvol6 4194304 105560 4056888 3% /tmp
/dev/vg00/lvol10 516096 80747 408822 16% /sysadm
/dev/vg00/lvol9 5865472 2493576 3345576 43% /opt
/dev/vg00/lvol5 6660096 1873792 4748944 28% /home
/dev/vg01/hprogram 491520 117083 351061 25% /home/program
/dev/vg01/hjpsc 409600 238780 160183 60% /home/jpsc
/dev/vg01/hcdp 6266880 386534 5513096 7% /home/cdp
/dev/vg01/hcaesar 1662976 212920 1359938 14% /home/caesar
/dev/vg01/caesar 33054720 18374980 14451500 56% /caesar
/dev/vg00/lvol4 106496 1133 98786 1% /.secure
sending to informix-list