Help with this performance of database

We have a web application on weblogic and backend is db 11 GR 2.
We also also running batch. people complain about the tasks performed too long - hrs.
Could someone comment on this AWR?

Thank you very much
              Snap Id      Snap Time      Sessions Curs/Sess
            --------- ------------------- -------- ---------
Begin Snap:     16147 28-Feb-13 16:00:08       815       3.4
  End Snap:     16148 28-Feb-13 17:00:21       871       3.6
   Elapsed:               60.22 (mins)
   DB Time:              148.96 (mins)

Cache Sizes                       Begin        End
~~~~~~~~~~~                  ---------- ----------
               Buffer Cache:     2,208M     2,208M  Std Block Size:         8K
           Shared Pool Size:     3,600M     3,600M      Log Buffer:    20,800K

Load Profile              Per Second    Per Transaction   Per Exec   Per Call
~~~~~~~~~~~~         ---------------    --------------- ---------- ----------
      DB Time(s):                2.5                0.1       0.00       0.00
       DB CPU(s):                2.1                0.1       0.00       0.00
       Redo size:          648,806.2           20,691.5
   Logical reads:          102,169.5            3,258.4
   Block changes:            5,538.9              176.6
  Physical reads:            9,146.3              291.7
 Physical writes:              182.8                5.8
      User calls:            1,049.4               33.5
          Parses:               38.5                1.2
     Hard parses:                1.5                0.1
W/A MB processed:                5.4                0.2
          Logons:                0.4                0.0
        Executes:              785.9               25.1
       Rollbacks:               12.7                0.4
    Transactions:               31.4

Top 5 Timed Foreground Events
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
                                                           Avg
                                                          wait   % DB
Event                                 Waits     Time(s)   (ms)   time Wait Class
------------------------------ ------------ ----------- ------ ------ ----------
DB CPU                                            7,403          82.8
log file sync                        63,357         954     15   10.7 Commit
db file sequential read           6,961,298         599      0    6.7 User I/O
direct path read                    326,419         230      1    2.6 User I/O
SQL*Net message from dblink           1,346         164    122    1.8 Network
Host CPU (CPUs:  160 Cores:   80 Sockets:   20)
~~~~~~~~         Load Average
               Begin       End     %User   %System      %WIO     %Idle
           --------- --------- --------- --------- --------- ---------
               27.61     28.95      10.5       6.0       0.0      83.6

Instance CPU
~~~~~~~~~~~~
              % of total CPU for Instance:       1.5
              % of busy  CPU for Instance:       8.9
  %DB time waiting for CPU - Resource Mgr:       0.0

Memory Statistics
~~~~~~~~~~~~~~~~~                       Begin          End
                  Host Mem (MB):    655,360.0    655,360.0
                   SGA use (MB):      6,144.0      6,144.0
                   PGA use (MB):      1,569.6      1,767.3
    % Host Mem used for SGA+PGA:         1.18         1.21


Statistic Name                                       Time (s) % of DB Time
------------------------------------------ ------------------ ------------
DB CPU                                                7,402.9         82.8
sql execute elapsed time                              6,861.1         76.8
PL/SQL execution elapsed time                           270.2          3.0
parse time elapsed                                      252.6          2.8
hard parse elapsed time                                 135.6          1.5
RMAN cpu time (backup/restore)                           61.2           .7
connection management call elapsed time                  20.0           .2
PL/SQL compilation elapsed time                           2.5           .0
inbound PL/SQL rpc elapsed time                           1.1           .0
hard parse (sharing criteria) elapsed time                0.3           .0
hard parse (bind mismatch) elapsed time                   0.1           .0
repeated bind elapsed time                                0.1           .0
sequence load elapsed time                                0.0           .0
DB time                                               8,937.8
background elapsed time                               3,823.8
background cpu time                                   1,067.8
          -------------------------------------------------------------

Operating System Statistics     


Statistic                                  Value        End Value
------------------------- ---------------------- ----------------
AVG_BUSY_TIME                             59,237
AVG_IDLE_TIME                            301,693
AVG_SYS_TIME                              21,399
AVG_USER_TIME                             37,656
BUSY_TIME                              9,507,251
IDLE_TIME                             48,300,375
SYS_TIME                               3,452,080
USER_TIME                              6,055,171
LOAD                                          28               29
OS_CPU_WAIT_TIME                         524,700
RSRC_MGR_CPU_WAIT_TIME                         0
VM_IN_BYTES                               65,536
VM_OUT_BYTES                                   0
PHYSICAL_MEMORY_BYTES            687,194,767,360
NUM_CPUS                                     160
NUM_CPU_CORES                                 80
NUM_CPU_SOCKETS                               20
TCP_RECEIVE_SIZE_DEFAULT               2,097,152
TCP_RECEIVE_SIZE_MAX                 335,544,320
TCP_SEND_SIZE_DEFAULT                  2,097,152
TCP_SEND_SIZE_MAX                    335,544,320
          -------------------------------------------------------------

Operating System Statistics - 

Snap Time           Load    %busy    %user     %sys    %idle  %iowait
--------------- -------- -------- -------- -------- -------- --------
28-Feb 16:00:08     27.6      N/A      N/A      N/A      N/A      N/A
28-Feb 17:00:21     28.9     16.4     10.5      6.0     83.6      0.0
          -------------------------------------------------------------


Foreground Wait Class           

-> Captured Time accounts for        107.0%  of Total DB time       8,937.75 (s)
-> Total FG Wait Time:             2,164.90 (s)  DB CPU time:       7,402.93 (s)

                                                                  Avg
                                      %Time       Total Wait     wait
Wait Class                      Waits -outs         Time (s)     (ms)  %DB time
-------------------- ---------------- ----- ---------------- -------- ---------
DB CPU                                                 7,403               82.8
Commit                         63,357     0              954       15      10.7
User I/O                    7,528,076     0              932        0      10.4
Network                     4,429,612     0              217        0       2.4
Application                     1,560     0               29       18       0.3
Other                         143,633     0               17        0       0.2
Concurrency                    50,758     0               10        0       0.1
Configuration                      71    14                4       62       0.0
System I/O                      7,295     0                2        0       0.0
Administrative                      2     0                0      136       0.0
          -------------------------------------------------------------

Foreground Wait Events         


                                                             Avg
                                        %Time Total Wait    wait    Waits   % DB
Event                             Waits -outs   Time (s)    (ms)     /txn   time
-------------------------- ------------ ----- ---------- ------- -------- ------
log file sync                    63,357     0        954      15      0.6   10.7
db file sequential read       6,961,298     0        599       0     61.4    6.7
direct path read                326,419     0        230       1      2.9    2.6
SQL*Net message from dblin        1,346     0        164     122      0.0    1.8
db file parallel read            75,275     0         51       1      0.7     .6
SQL*Net more data to clien    1,468,920     0         45       0     13.0     .5
enq: KO - fast object chec          467     0         20      43      0.0     .2
direct path write                 1,890     0         19      10      0.0     .2


Background Wait Events         

                                                             Avg
                                        %Time Total Wait    wait    Waits   % bg
Event                             Waits -outs   Time (s)    (ms)     /txn   time
-------------------------- ------------ ----- ---------- ------- -------- ------
log file parallel write          71,426     0        393       5      0.6   10.3
db file parallel write           31,764     0        223       7      0.3    5.8
Backup: MML create a backu            4     0        163   40660      0.0    4.3
Backup: MML write backup p       14,976     0        153      10      0.1    4.0
control file parallel writ        5,276     0         81      15      0.0    2.1
Backup: MML commit backup             4     0         50   12598      0.0    1.3
log file sequential read            530     0         21      39      0.0     .5
db file async I/O submit         10,968     0         17       2      0.1     .5
db file sequential read           4,562     0         17       4      0.0     .4

If you actually 160 hearts, most of the drivers is inactive (so it's not one of the dozens of databases on the same server, all the competitor for the same 160 cores), you have the Enterprise edition (which I believe is required for a parallel query, but it is also a sine qua non for the Performance and Tuning Pack (, which would be necessary for the AWR report you posted) and batch processing are parallelizable, Yes, that would certainly improve the performance of the batch. If batches are parallelizazble will depend on how they are built. If they do operations based on a game, for example, Oracle can often use a parallel query to significantly improve performance. If you have a lot of line of PL/SQL, it'll take more work to run in parallel.

Justin

Tags: Database

Similar Questions

Maybe you are looking for