Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> Help on Oraperf YAPP analyzer

Help on Oraperf YAPP analyzer

From: <robson_at_medidata.com.br>
Date: Mon, 7 Jun 2004 15:46:05 -0300
Message-ID: <OF781811DA.AC4248F4-ON03256EAC.0065BE1F@medidata.com.br>


Hi

This is my fitst post to the list.
I used the oraperf analyzer on some 8.1.7.4 statspack reports and found different results from the ones I got last year. Comparing results for the same statspack reports I could locate the source of the difference:
On the response time calculation, now but not before, the wait times are being multiplied by 100.
This gives one amount of wait time two orders of magnitude greater than the cpu times, and biases the response times as being dominated by huge wait times.
Does someone noted this behavior change? Any help is appreciated

Below are the excerpts of the YAPP oraperf output from the same statspack report:

Old report

Breakdown of Response Time

|------------+------+----------+---------+----------+------------|

| | Time | Percentag| Per | Per User | Per |
| | | e | Execute | Call | Transaction|
|------------+------+----------+---------+----------+------------|
| Response | | | | | |
| Time | | | | | |
|------------+------+----------+---------+----------+------------|
| Response | 77712| 100.00%| 0.61| 0.34| 15.68|
| Time | 9| | | | |
|------------+------+----------+---------+----------+------------|
| CPU Time | 42197| 54.30%| 0.33| 0.19| 8.51|
| | 6| | | | |
|------------+------+----------+---------+----------+------------|
| Wait Time | 35515| 45.70%| 0.28| 0.16| 7.17|
| | 3| | | | |
|------------+------+----------+---------+----------+------------| New report I think here the wait time is inflated by a factor times 100: Breakdown of Response Time |------------+--------+----------+---------+----------+------------|
| | Time | Percentag| Per | Per User | Per |
| | | e | Execute | Call | Transaction|
|------------+--------+----------+---------+----------+------------|
| Response | | | | | |
| Time | | | | | |
|------------+--------+----------+---------+----------+------------|
| Response | 3593727| 100.00%| 28.17| 15.87| 725.05|
| Time | 6| | | | |
|------------+--------+----------+---------+----------+------------|
| CPU Time | 421976| 1.17%| 0.33| 0.19| 8.51|
|------------+--------+----------+---------+----------+------------|
| Wait Time | 3551530| 98.83%| 27.84| 15.69| 716.54|
| | 0| | | | |
|------------+--------+----------+---------+----------+------------|

If you look on the wait time break down now all times are times 100: Breakdown of Wait Time

The wait for the foreground sessions can be broken down in the following wait events (in order of wait time):

|---------------------+--------+----------+--------+---------+----------+------------|

| Event | Time | Percentag| Avg. | Per | Per User | Per |
| | | e | Wait | Execute | Call | Transaction|
|---------------------+--------+----------+--------+---------+----------+------------|
| Wait Time | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| log file sync | 2195490| 61.82%| 97.69| 17.21| 9.70| 442.95|
| | 0| | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| db file sequential | 9333600| 26.28%| 3.34| 7.32| 4.12| 188.31|
| read | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| direct path write | 1866800| 5.26%| 1026.28| 1.46| 0.82| 37.66|
|---------------------+--------+----------+--------+---------+----------+------------|
| db file scattered | 811000| 2.28%| 17.67| 0.64| 0.36| 16.36|
| read | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| SQL*Net more data to| 345200| 0.97%| 52.60| 0.27| 0.15| 6.96|
| client | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| direct path read | 323700| 0.91%| 74.83| 0.25| 0.14| 6.53|
|---------------------+--------+----------+--------+---------+----------+------------|
| SQL*Net message from| 252400| 0.71%| 32.61| 0.20| 0.11| 5.09|
| dblink | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| latch free | 196600| 0.55%| 26.68| 0.15| 0.09| 3.97|
|---------------------+--------+----------+--------+---------+----------+------------|
| control file | 124400| 0.35%| 51.68| 0.10| 0.05| 2.51|
| parallel write | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| SQL*Net break/reset | 77600| 0.22%| 2.72| 0.06| 0.03| 1.57|
| to clien | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| SQL*Net more data | 67900| 0.19%| 2.82| 0.05| 0.03| 1.37|
| from dblin | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| log file sequential | 58200| 0.16%| 3.03| 0.05| 0.03| 1.17|
| read | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| SQL*Net message to | 53500| 0.15%| 0.02| 0.04| 0.02| 1.08|
| client | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| file open | 22800| 0.06%| 6.71| 0.02| 0.01| 0.46|
|---------------------+--------+----------+--------+---------+----------+------------|
| enqueue | 7500| 0.02%| 250.00| 0.01| 0.00| 0.15|
|---------------------+--------+----------+--------+---------+----------+------------|
| log file switch | 4000| 0.01%| 571.43| 0.00| 0.00| 0.08|
| completion | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| buffer busy waits | 3900| 0.01%| 9.22| 0.00| 0.00| 0.08|
|---------------------+--------+----------+--------+---------+----------+------------|
| refresh controlfile | 2700| 0.01%| 2.10| 0.00| 0.00| 0.05|
| command | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| control file | 2700| 0.01%| 0.38| 0.00| 0.00| 0.05|
| sequential read | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| SQL*Net more data | 2600| 0.01%| 0.45| 0.00| 0.00| 0.05|
| from clien | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| library cache pin | 1600| 0.00%| 800.00| 0.00| 0.00| 0.03|
|---------------------+--------+----------+--------+---------+----------+------------|
| direct path read | 700| 0.00%| 1.55| 0.00| 0.00| 0.01|
| (lob) | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| log file single | 600| 0.00%| 100.00| 0.00| 0.00| 0.01|
| write | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| SQL*Net message to | 300| 0.00%| 0.04| 0.00| 0.00| 0.01|
| dblink | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|
| file identify | 100| 0.00%| 6.67| 0.00| 0.00| 0.00|
|---------------------+--------+----------+--------+---------+----------+------------|
| SQL*Net break/reset | 0| 0.00%| 0.00| 0.00| 0.00| 0.00|
| to dblin | | | | | | |
|---------------------+--------+----------+--------+---------+----------+------------|

Robson Gomes

Medidata Informática S.A



Please see the official ORACLE-L FAQ: http://www.orafaq.com

To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line.
--
Archives are at http://www.freelists.org/archives/oracle-l/
FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------
Received on Mon Jun 07 2004 - 13:44:08 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US