From: Abdellah Djebli (abdellah.djebli@SICKKIDS.CA)
Date: Mon Dec 29 2003 - 14:19:33 EST
Hi All,
We have a system (6xCPU and 12GB RAM) running DB2 connected to an IBM ESS
through two fiber SAN connections.
My question i, why is it that Wait% column always shows 25-30 % where there
is no IO/memory or CPU bottleneck.
What are the CPUs waiting for?
Thanks in advance.
Abdu
Here is an output of vmstat 1 20
kthr memory page faults cpu
----- ----------- ------------------------ ------------ -----------
r b avm fre re pi po fr sr cy in sy cs us sy id wa
3 1 601868 1086361 0 0 0 78 182 0 1013 2964 1867 15 1 80 3
1 2 601872 1086325 0 0 0 0 0 0 1782 5676 4217 19 3 47 32
1 4 601872 1086256 0 0 0 0 0 0 1936 10431 7614 29 4 41 26
0 1 601872 1086223 0 0 0 0 0 0 1481 4182 2966 0 2 78 21
0 1 601872 1086209 0 0 0 0 0 0 1500 6300 4921 11 2 69 18
1 1 601872 1086300 0 0 0 0 0 0 1563 4618 3248 5 2 70 23
0 2 601872 1086424 0 0 0 0 0 0 2568 7020 6807 7 3 57 32
1 5 601879 1086641 0 0 0 0 0 0 2066 7936 5657 5 4 59 31
1 2 601879 1086737 0 0 0 0 0 0 1590 6203 4937 11 3 57 30
1 3 601879 1086928 0 0 0 0 0 0 1645 5750 3425 3 2 64 31
1 3 601893 1086913 0 0 0 0 0 0 1764 7277 3618 4 2 63 31
0 2 601879 1086936 0 0 0 0 0 0 1953 7266 5748 13 4 59 25
0 2 601879 1087053 0 0 0 0 0 0 1547 5753 3051 7 3 62 28
3 1 601879 1086912 0 0 0 0 0 0 2150 7022 4256 13 5 51 32
0 2 601879 1086967 0 0 0 0 0 0 1441 4229 2847 2 2 72 24
0 2 601879 1086979 0 0 0 0 0 0 1280 3819 2411 1 2 77 21
0 1 601879 1086960 0 0 0 0 0 0 1332 3871 2574 1 2 78 20
0 1 601879 1086993 0 0 0 0 0 0 1321 3946 2622 2 2 77 19
1 1 601879 1087064 0 0 0 0 0 0 1554 6473 5221 19 2 60 19
3 0 601879 1087149 0 0 0 0 0 0 1456 5503 3235 16 3 61 20
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 22:17:25 EDT