Shah ., December 01, 2016 - 8:09 pm UTC
Thanks Chris,
Well I took the same piece of data into another environment.. ran the same steps... it came out in around 15 mins...
December 02, 2016 - 2:37 am UTC
Glad we could help
Shah ., December 01, 2016 - 8:09 pm UTC
Thanks Chris,
Well I took the same piece of data into another environment.. ran the same steps... it came out in around 15 mins...
Shah ., December 01, 2016 - 8:09 pm UTC
Thanks Chris,
Well I took the same piece of data into another environment.. ran the same steps... it came out in around 15 mins...
investigate on sql qry
A reader, December 02, 2016 - 8:51 am UTC
"If you have AWR or statspack data available, check that to see what the difference between the executions was. This will clue you into the real problem and help you figure out what do to. "
i have a qry (merge into GTT using a select) :
how look at tre AWR report to find out the reason of slowness ?
from the report :
CPU Time (s) : 1,300.10
Executions : 2
UIO per Exec (s) : 2,977.82
CPU per Exec (s) : 650.05
%Total : 8.67
Elapsed Time (s) : 9,365.86
%CPU : 13.88
%IO : 63.59
SQL Id : 8urny8p02u49a
Gets per Exec : 82,103,589.50
%Total : 5.86
Reads per Exec : 79,046,315.00
UnOptimized Read Reqs : 11,420,196
Physical Read Reqs : 11,420,196
so?
A reader, December 02, 2016 - 10:26 am UTC
Tkx chris for replying
i get the plans and see the link
but... what tells me WHY the execution has taken 2 hours instead of minutes before ?
December 02, 2016 - 2:40 pm UTC
If you find you have two different execution plans, then that's almost certainly your "why".
If it's the same plan for both runs you need to do some investigating. e.g. Was it waiting on something one time but not the other? Did it do lots more physical IO? Was IO much slower for some reason?
You need to compare the periods to find out what's different.