IO on tempdbs
Posted in 2000
Topics: Storage & Space Management, Versions, Editions & End-of-Life
Hi,
SUN E10K Server
IDS 7.30.UC5
4GB Memory
10 CPUs
From the following output, there's a high IO on the tempdb01. Can I create
another chunk (tempdb02) to reduce the IO on existing chunk???
onstat -g iofAIO global files:
gfd pathname totalops dskread dskwrite io/s
3 rootdb01 5917 5402 515 0.2
4 llogdb01 150351 59635 90716 4.0
5 tempdb01 8344590 6790797 1553793 220.0
6 datdb01 1534681 1443097 91584 40.5
7 datdb02 559576 493398 66178 14.8
8 datdb03 2440094 2400980 39114 64.3
9 datdb04 830796 797420 33376 21.9
10 datdb05 4041916 3972418 69498 106.6
11 datdb06 1440962 638299 802663 38.0
12 datdb07 741770 426267 315503 19.6
13 datdb08 3037042 2967223 69819 80.1
14 datdb09 1204842 1127972 76870 31.8
15 datdb10 1197620 1118538 79082 31.6
16 dwdb01 374763 160166 214597 9.9
17 dwdb02 967169 529044 438125 25.5
18 dwdb03 1092875 886876 205999 28.8
19 dwdb04 978854 417472 561382 25.8
20 cycdb01 11774 11043 731 0.3
21 dwdb05 247961 54800 193161 6.5
22 datdb11 2329385 2244621 84764 61.4
23 llogdb02 56762 21355 35407 1.5
24 datdb12 1343574 1197043 146531 35.4
25 phydbs 66302 10 66292 1.7
26 datdb13 4 4 0 0.0
27 datdb14 4 4 0 0.0
28 datdb15 4 4 0 0.0
29 dwdb06 325315 65840 259475 8.6
30 dwdb07 4 4 0 0.0
31 dwdb08 4 4 0 0.0
Regards,
Jason
Create 3 temporary dbspaces (using the -t option)
and have one chunk per temporary dbspace.
Also consider moving to 7.31.UC5.
JasonYLPang@pg.SLR.com wrote in message <8qn3ae$9mn$1@news.xmission.com>...
>
>Hi,
>
>SUN E10K Server
>IDS 7.30.UC5
>4GB Memory
>10 CPUs
>
>From the following output, there's a high IO on the tempdb01. Can I create
>another chunk (tempdb02) to reduce the IO on existing chunk???
>
>onstat -g iof>AIO global files:
>gfd pathname totalops dskread dskwrite io/s
> 3 rootdb01 5917 5402 515 0.2
> 4 llogdb01 150351 59635 90716 4.0
> 5 tempdb01 8344590 6790797 1553793 220.0
> 6 datdb01 1534681 1443097 91584 40.5
> 7 datdb02 559576 493398 66178 14.8
> 8 datdb03 2440094 2400980 39114 64.3
> 9 datdb04 830796 797420 33376 21.9
> 10 datdb05 4041916 3972418 69498 106.6
> 11 datdb06 1440962 638299 802663 38.0
> 12 datdb07 741770 426267 315503 19.6
> 13 datdb08 3037042 2967223 69819 80.1
> 14 datdb09 1204842 1127972 76870 31.8
> 15 datdb10 1197620 1118538 79082 31.6
> 16 dwdb01 374763 160166 214597 9.9
> 17 dwdb02 967169 529044 438125 25.5
> 18 dwdb03 1092875 886876 205999 28.8
> 19 dwdb04 978854 417472 561382 25.8
> 20 cycdb01 11774 11043 731 0.3
> 21 dwdb05 247961 54800 193161 6.5
> 22 datdb11 2329385 2244621 84764 61.4
> 23 llogdb02 56762 21355 35407 1.5
> 24 datdb12 1343574 1197043 146531 35.4
> 25 phydbs 66302 10 66292 1.7
> 26 datdb13 4 4 0 0.0
> 27 datdb14 4 4 0 0.0
> 28 datdb15 4 4 0 0.0
> 29 dwdb06 325315 65840 259475 8.6
> 30 dwdb07 4 4 0 0.0
> 31 dwdb08 4 4 0 0.0
>
>Regards,
>Jason
I've had some really good luck striping the tempdbs across multiple drives
at the OS level.
<JasonYLPang@pg.SLR.com> wrote in message
news:8qn3ae$9mn$1@news.xmission.com...
>
> Hi,
>
> SUN E10K Server
> IDS 7.30.UC5
> 4GB Memory
> 10 CPUs
>
> From the following output, there's a high IO on the tempdb01. Can I create
> another chunk (tempdb02) to reduce the IO on existing chunk???
>
> onstat -g iof> AIO global files:
> gfd pathname totalops dskread dskwrite io/s
> 3 rootdb01 5917 5402 515 0.2
> 4 llogdb01 150351 59635 90716 4.0
> 5 tempdb01 8344590 6790797 1553793 220.0
> 6 datdb01 1534681 1443097 91584 40.5
> 7 datdb02 559576 493398 66178 14.8
> 8 datdb03 2440094 2400980 39114 64.3
> 9 datdb04 830796 797420 33376 21.9
> 10 datdb05 4041916 3972418 69498 106.6
> 11 datdb06 1440962 638299 802663 38.0
> 12 datdb07 741770 426267 315503 19.6
> 13 datdb08 3037042 2967223 69819 80.1
> 14 datdb09 1204842 1127972 76870 31.8
> 15 datdb10 1197620 1118538 79082 31.6
> 16 dwdb01 374763 160166 214597 9.9
> 17 dwdb02 967169 529044 438125 25.5
> 18 dwdb03 1092875 886876 205999 28.8
> 19 dwdb04 978854 417472 561382 25.8
> 20 cycdb01 11774 11043 731 0.3
> 21 dwdb05 247961 54800 193161 6.5
> 22 datdb11 2329385 2244621 84764 61.4
> 23 llogdb02 56762 21355 35407 1.5
> 24 datdb12 1343574 1197043 146531 35.4
> 25 phydbs 66302 10 66292 1.7
> 26 datdb13 4 4 0 0.0
> 27 datdb14 4 4 0 0.0
> 28 datdb15 4 4 0 0.0
> 29 dwdb06 325315 65840 259475 8.6
> 30 dwdb07 4 4 0 0.0
> 31 dwdb08 4 4 0 0.0
>
> Regards,
> Jason
Related threads
- Posting from the Informix-list
- Migrating from IDS 9.40.UC6 to 11.50.UC3
- Ip for a network session
- questions onstat -g