The 'w' column in the vmstat output below... has something to do with demoted IO. When Oracle redo logs with default block size of 512 bytes are put on a JFS2 cio mounted filesystem, for example, with the default 4096 byte filesystem block size.
Jaqui Lynch referenced this vmstat option in her March 2014 presentation, page 41.
https://www.ibm.com/developerworks/community/wikis/form/anonymous/api/wiki/61ad9cf2-c6a3-4d2c-b779-61ff0266d32a/page/1cb956e8-4160-4bea-a956-e51490c2b920/attachment/52ed9996-7561-42e8-a446-09f5f1414521/media/VUG-AIXPerformanceTuning-Part2-mar2414.pdf
But I haven't found an explanation of what the conditions are for the w column to count a thread. Not yet, anyway. The second variation below is used in perfpmr memdetails.sh script.
# vmstat -IW
System configuration: lcpu=12 mem=16384MB ent=1.00
kthr memory page faults cpu
----------- ----------- ------------------------ ------------ -----------------------
r b p w avm fre fi fo pi po fr sr in sy cs us sy id wa pc ec
1 1 0 0 1835258 1779800 0 2 0 0 0 0 24 1991 1625 0 0 99 0 0.01 1.0
# vmstat -W -h -t -w -I 2 2
System configuration: lcpu=12 mem=16384MB ent=1.00
kthr memory page faults cpu hypv-page time
--------------- --------------------- ------------------------------------ ------------------ ----------------------- ------------------------- --------
r b p w avm fre fi fo pi po fr sr in sy cs us sy id wa pc ec hpi hpit pmem loan hr mi se
2 0 0 0 1824787 1788933 0 15 0 0 0 0 11 2874 1698 3 3 94 0 0.10 10.0 0 0 16.00 0.00 16:34:26
2 0 0 0 1824802 1788916 0 0 0 0 0 0 6 1793 1680 1 1 98 0 0.05 4.9 0 0 16.00 0.00 16:34:28
No comments:
Post a Comment