DB Time spent nowhere in AWR [message #683286] |
Thu, 24 December 2020 05:40 |
|
piripicchio
Messages: 20 Registered: April 2018 Location: Rome
|
Junior Member |
|
|
Hi all,
while investigating an high response time per transaction, I came across an unusual (at least, for me) situation in which I can't find where the DB time is spent; for instance, below I copied the time model section from the AWR where you can see that no components seem to be contributing to the total DB Time. The same if we look at other figures in the report, like wait events: every figure is a fractional percent of the DB Time. Does anybody have any idea where I should loot at to find where this db time is spent?
Thank you in advance.
Time Model Statistics
% of % of Total
Statistic Name Time (s) DB Time CPU Time
------------------------------------------ ------------------ ------- ----------
sql execute elapsed time 935.0 .9
DB CPU 607.5 .6 64.7
parse time elapsed 25.8 .0
hard parse elapsed time 23.3 .0
PL/SQL execution elapsed time 16.2 .0
hard parse (sharing criteria) elapsed time 6.8 .0
hard parse (bind mismatch) elapsed time 2.0 .0
connection management call elapsed time 1.3 .0
PL/SQL compilation elapsed time 0.2 .0
sequence load elapsed time 0.1 .0
failed parse elapsed time 0.0 .0
repeated bind elapsed time 0.0 .0
DB time 106,063.7
background elapsed time 335.3
background cpu time 331.0 35.3
total CPU time 938.5
------------------------------------------------------
|
|
|
Re: DB Time spent nowhere in AWR [message #683287 is a reply to message #683286] |
Thu, 24 December 2020 07:46 |
John Watson
Messages: 8963 Registered: January 2010 Location: Global Village
|
Senior Member |
|
|
DB Time arithmetic is often a bit odd. Operating system, platform, and virtualization can really confuse it. For example, AIX accounts for IO time very differently from Linux. Different virtualization methods may mean you have to interpret the numbers very differently. If you post the complete report, it might be possible for someone to comment.
|
|
|