How to reduce the cloning time if using cold backup?

Hi,
We are using EBS r12 (12.0.6) with database 10.2.0.3 in Linux Redhat 4 32-bit envoirnment.
Our datbase size is around 480 GB and we are facing the issues to provide the clone to my consultants for meet the target timelines in given time.
(Source:)---- Production Dell R900 Server machine having 32-GB RAM and 8 Quard Core CPU's.
(Target:)---- Clone Dell 2950 Server machine having 16-GB RAM and 4-Quard Core CPU's.
Currectly we are taking cold backup like:
1: Auto shutdown EBS r12 database nighthy at 12:00 AM daily and compress the backup using tar utility in linux. This process takes 6:00 hours aprox:
2: After that we move compressed file to clone machine and then uncompressed it, and this process takes 5:00 hours aprox.
3: And performe standard cloning steps.
Questions:
1: How to reduce time of this backup process?
3: Is there any other way to reduce the cloning process.
2: What type of backup oracle recommended to their customers for this type of process?
Thanks.

1: How to reduce time of this backup process?Without using third party tools, it might be hard to tune the timing of compressing/uncompressing the file.
Have you tried to use scp command? This would help if your network throughput is acceptable.
3: Is there any other way to reduce the cloning process.Since the main issue you have with the copy, then you might copy the files remotely from the source to the target, or using any other storage/backup tools (like file system snapshot).
2: What type of backup oracle recommended to their customers for this type of process?Oracle does not recommend any type of backup as the tools used are not Oracle products.
Thanks,
Hussein

Similar Messages

How to reduce the fetch time of this sql?

Here is the SQL, three-table join and joining conditions are:
ms_ets_cntrl.ims_cntrt_oid=ims_alctn.ims_alctn_oid
ims_alctn.ims_trde_oid=ims_trde.ims_trde_oid
SELECT 'MCH' Type, ims_ets_cntrl.STTS tp_stts, count(*) Count FROM ims_ets_cntrl where ims_ets_cntrl.ims_cntrt_oid in
(select ims_alctn.ims_alctn_oid FROM ims_alctn, ( select ims_trde.ims_trde_oid from ims_trde
where ( IMS_TRDE.IMS_TRDE_RCPT_DTTM >= TO_DATE('10/29/2009 00:00', 'MM/DD/YYYY HH24:MI') AND IMS_TRDE.IMS_TRDE_RCPT_DTTM <= (TO_DATE('11/5/2009 23:59', 'MM/DD/YYYY HH24:MI')) )
AND (IMS_TRDE.GRS_TRX_TYPE IN ('INJECTION','WITHDRAWAL','PAYMENT') OR IMS_TRDE.SSC_INVST_TYPE = 'FC' AND 1=1 and IMS_TRDE.SERVICE_TYPE='FS' )) TRDE
where IMS_ALCTN.IMS_TRDE_OID=TRDE.IMS_TRDE_OID) and ims_ets_cntrl.outbnd_dest = 'ETD' group by ims_ets_cntrl.STTSOptimizer and related parameter info:
SQL> show parameter optimizer
NAME                                 TYPE        VALUE
optimizer_dynamic_sampling           integer     1
optimizer_features_enable            string      9.2.0
optimizer_index_caching              integer     0
optimizer_index_cost_adj             integer     100
optimizer_max_permutations           integer     2000
optimizer_mode                       string      CHOOSE
SQL>select pname, pval1, pval2 from sys.aux_stats$ where sname='SYSSTATS_INFO';
DSTART          11-16-2009 10:23
DSTOP          11-16-2009 10:23
FLAGS     1
STATUS          NOWORKLOADHere is autotrace output:
Execution Plan
   0      SELECT STATEMENT Optimizer=CHOOSE
   1    0   SORT (GROUP BY)
   2    1     VIEW
   3    2       SORT (UNIQUE)
   4    3         TABLE ACCESS (BY INDEX ROWID) OF 'IMS_ETS_CNTRL'
   5    4           NESTED LOOPS
   6    5             NESTED LOOPS
   7    6               TABLE ACCESS (BY INDEX ROWID) OF 'IMS_TRDE'
   8    7                 INDEX (RANGE SCAN) OF 'IMS_TRDE_INDX4' (NON- UNIQUE)
   9    6               TABLE ACCESS (BY INDEX ROWID) OF 'IMS_ALCTN'
10    9                 INDEX (RANGE SCAN) OF 'IMS_ALCTN_INDX1' (NON -UNIQUE)
11    5             INDEX (RANGE SCAN) OF 'IMS_ETS_CNTRL_INDX1' (NON -UNIQUE)
Statistics
          0 recursive calls
          0 db block gets
     244608 consistent gets
      58856 physical reads
          0 redo size
        497 bytes sent via SQL*Net to client
        499 bytes received via SQL*Net from client
          2 SQL*Net roundtrips to/from client
          2 sorts (memory)
          0 sorts (disk)
          1 rows processedHere is TKPROF output:
call     count       cpu    elapsed       disk      query    current        rows
Parse        1      0.00       0.00          0          0          0           0
Execute      1      0.00       0.00          0          0          0           0
Fetch        2      4.85     129.72      53863     244608          0           1
total        4      4.85     129.72      53863     244608          0           1
Misses in library cache during parse: 1
Optimizer goal: CHOOSE
Parsing user id: 63
Rows     Row Source Operation
      1 SORT GROUP BY
12972   VIEW
12972    SORT UNIQUE
12972     TABLE ACCESS BY INDEX ROWID IMS_ETS_CNTRL
46236      NESTED LOOPS
19134       NESTED LOOPS
19744        TABLE ACCESS BY INDEX ROWID IMS_TRDE
176922         INDEX RANGE SCAN IMS_TRDE_INDX4 (object id 34099)
19134        TABLE ACCESS BY INDEX ROWID IMS_ALCTN
19134         INDEX RANGE SCAN IMS_ALCTN_INDX1 (object id 34094)
27101       INDEX RANGE SCAN IMS_ETS_CNTRL_INDX1 (object id 34101)
********************************************************************************Explain plan output:
PLAN_TABLE_OUTPUT
| Id | Operation                         | Name                | Rows | Bytes | Cost |
|   0 | SELECT STATEMENT                  |                      |       |       |       |
|   1 | SORT GROUP BY                    |                      |       |       |       |
|   2 |   VIEW                            |                      |       |       |       |
|   3 |    SORT UNIQUE                    |                      |       |       |       |
|* 4 |     TABLE ACCESS BY INDEX ROWID   | IMS_ETS_CNTRL        |       |       |       |
|   5 |      NESTED LOOPS                 |                      |       |       |       |
|   6 |       NESTED LOOPS                |                      |       |       |       |
|* 7 |        TABLE ACCESS BY INDEX ROWID| IMS_TRDE             |       |       |       |
|* 8 |         INDEX RANGE SCAN          | IMS_TRDE_INDX4       |       |       |       |
|   9 |        TABLE ACCESS BY INDEX ROWID| IMS_ALCTN            |       |       |       |
|* 10 |         INDEX RANGE SCAN          | IMS_ALCTN_INDX1      |       |       |       |
|* 11 |       INDEX RANGE SCAN            | IMS_ETS_CNTRL_INDX1 |       |       |       |
Predicate Information (identified by operation id):
   4 - filter("IMS_ETS_CNTRL"."OUTBND_DEST"='ETD')
   7 - filter("IMS_TRDE"."GRS_TRX_TYPE"='INJECTION' OR "IMS_TRDE"."GRS_TRX_TYPE"='WITHD
              RAWAL' OR "IMS_TRDE"."GRS_TRX_TYPE"='PAYMENT' OR "IMS_TRDE"."SSC_INVST_TY
              PE"='FC' AND "IMS_TRDE"."SERVICE_TYPE"='FS')
   8 - access("IMS_TRDE"."IMS_TRDE_RCPT_DTTM">=TO_DATE('2009-10-29 00:00:00', 'yyyy-mm-
              dd hh24:mi:ss') AND "IMS_TRDE"."IMS_TRDE_RCPT_DTTM"<=TO_DATE('2009-11-05
              23:59:00', 'yyyy-mm-dd hh24:mi:ss')
10 - access("IMS_ALCTN"."IMS_TRDE_OID"="IMS_TRDE"."IMS_TRDE_OID")
11 - access("IMS_ETS_CNTRL"."IMS_CNTRT_OID"="IMS_ALCTN"."IMS_ALCTN_OID")
Note: rule based optimizationCould you please help tune this sql?
How can I reduce the elapsed time? How can I reduce query read?
If there is any other info that you need, please let me know!
thank you very much!

What exactly is this logic meant to do?
AND    (ims_trde.grs_trx_type IN ('INJECTION', 'WITHDRAWAL', 'PAYMENT')
        OR ims_trde.ssc_invst_type = 'FC'
        AND ims_trde.service_type = 'FS')is that really:
AND    (ims_trde.grs_trx_type IN ('INJECTION', 'WITHDRAWAL', 'PAYMENT')
        OR ims_trde.ssc_invst_type = 'FC')
AND    ims_trde.service_type = 'FS'or is it maybe:
AND   (ims_trde.grs_trx_type IN ('INJECTION', 'WITHDRAWAL', 'PAYMENT')
       OR (ims_trde.ssc_invst_type = 'FC'
           AND ims_trde.service_type = 'FS'))?

How to reduce the elapsed time on this query.

Oracle10.2.0.5 on AIX6.1.
Here is the awrsql report of a query for a one hour window.
Here UNBILLED has 120million records which is partitioned and subpartitioned. It has 10 partitions and 312 subpartitions. One subpartition has 10M records, 10 others have 1M each and 270 subpartitions have 360K records each. And 31 subpartitions are empty.
And BILLED has 300K records.
Looking at report you can see that out of 984 seconds, 958 is used for IO. Machine is PowerPC_POWER5 with 8CPU and disk storage on EMC.
Stat Name                                Statement   Per Execution % Snap
Elapsed Time (ms)                           984,327        5,561.2    28.6
CPU Time (ms)                                40,760          230.3    29.8
Executions                                      177            N/A     N/A
Buffer Gets                               1,923,456       10,867.0    46.9
Disk Reads                                  203,832        1,151.6    25.1
Parse Calls                                       0            0.0     0.0
Rows                                        106,200          600.0     N/A
User I/O Wait Time (ms)                     958,766            N/A     N/A
Cluster Wait Time (ms)                            0            N/A     N/A
Application Wait Time (ms)                        0            N/A     N/A
Concurrency Wait Time (ms)                        0            N/A     N/A
Invalidations                                     0            N/A     N/A
Version Count                                     1            N/A     N/A
Sharable Mem(KB)                                 31            N/A     N/A
SQL_ID akg45a750sh0u
DELETE UNBILLED WHERE (MSG_ID, MSG_ID2,SPLIT_ROW_NUM) IN (SELECT MSG_ID,MSG_ID2, SPLIT_ROW_NUM FROM BILLED
WHERE BILL_REF_NO = :B2 AND BILL_REF_RESETS = :B1 )
Plan hash value: 3879210699
| Id | Operation                            | Name                          | Rows | Bytes | Cost (%CPU)| Pstart| Pstop |
|   0 | DELETE STATEMENT                     |                               |       |       |     1 (100)|       |       |
|   1 | DELETE                              | UNBILLED                      |       |       |            |       |       |
|   2 |   NESTED LOOPS                       |                               |     1 |   112 |     0   (0)|       |       |
|   3 |    TABLE ACCESS BY GLOBAL INDEX ROWID| BILLED                        |     1 |    65 |     0   (0)|     1 |     1 |
|   4 |     INDEX RANGE SCAN                 | BILLED_XCB_BILL_REF_TRANS     |     1 |       |     0   (0)|       |       |
|   5 |    INDEX UNIQUE SCAN                 | UNBILLED_PK                   |     1 |    47 |     0   (0)|       |       |
----------------------------------------------------------------------------------------------------------------------Is there way to improve the elapsed time( currently 5,561.2 msec per execution) of the query?
Thanks for your time...

user1014212 wrote:
You're giving very interresting pointers.. Especially the one using the GTT table... Thanks...
I was also looking from a DBA point of view, say increasing the DB_CACHE_SIZE from 5GB to 8GB as the machine has almost 32GB memory out of which 13GB is free memory. Because the timed event is "db file sequential read" which takes 95% of Total Call time.
select * from V_$SGA_TARGET_ADVICE;
post SQL & FORMATTED results using tags

After the test DB created using cold backup..things to do..

Dear All,
I had done Physical cold backup cloning to create the test DB using Production DB.
After that ,Is anything need to do before give to the application team?
For say,
index recreation,
index re-build,
valid the invalid objects
Please help me and input your suggestions.
Thanks in Advance,
Regards,
DB.

All objects should be fine – no need to rebuild/recompile anything.
If tables in the production DB contain references to production HTMLs, production server names, production usernames/passwords, etc, they need to be changed or deleted. The DB password should be changed.
If the production DB has DB links to other production DBs, those DB links have to be redefined.
If any of the data in the production DB should not be visible to the developers, it should be masked.
Iordan Iotzov

How to reduce the Logout time in IDM

Hi Everyone,
Help me in Solving the issue.
When i click on the logout button of the end user interface in sun idm, it is taking 20 secs to come back to the login page (login.jsp) which is causing very much delay in the process. I want to reduce this time. Dont know exactly what is the backend process.
Kindly help me if you know any solution for the same.
Thank you in advance.

Srikanth,
It would be better to use a database table(s) to store your audit trail information than an xml file, which could grow beyond a manageable size and then require housekeeping efforts and file rolling, etc. It would also be much easier to query user related information out of the logs, pareto charts for failed vs. successful login attempts, etc. if the data was stored in a database.
Have you considered modifying the Relogin.jsp page link in the portal's sub-menu bar? How about making your own version of this web page and changing the link pointer? Keep in mind that whatever result you come up with here will need a significant update for NetWeaver UME compatibility in version 12.0.
Regards,
Jeremy

Oracle 10g: How to reduce the select * time

Hello,
We have 10 million entries in our database (Oracle 10g in windows 32 bit machine.)
The execution of 'select * ....' takes 3 to 4 hour. Is there any way to reduce this time.
Is any tool available which can read the oracle export data and produce the output in text file format.
or any idea ?
Thanks
With Regards
Hemant.

hem_kec wrote:
Hello EdStevens
Is that 3 to 4 hours scrolling data past your screen?Answer: The Oracle is taking 3-4 hr to produce the output.
OK, let me try again. Where is the output being directed? To the screen? To a file?
The reason I ask is that often people will say "It takes n minutes to run this query" when in fact Oracle is producing the result set in a few seconds and what is taking n minutes is to run the results past the screen.
You should take a statspack report while the query is running and see where it is spending its time.
>
That's a different problem. I assume by "export data" you mean a .dmp file created by exp or expdp? If so what do you hope to achieve by outputting it in text format? What is the business problem to be solved?Answer: Since customer want to read all 10 milion entries. so we are think if we can dump (Oracle export) the data and using some tool (not Oracle) to read the data so there is no wait for the customer.As stated, a dmp file is oracle proprietary binary file that exists solely to transport data across platforms/versions. It is not suitable for your purpose. You are far better off finding where the current query is spending its time than looking for some kludge. 10 million rows of data is still 10 million rows of data. Do you think extracting it from the db, storeing it in some other format, and having some other tool read it off of disk in that format is going to be faster than simply selecting it from the db -- asking the db to do what it was designed to do?
>
>
Thanks
With Regards
Hemant.

How to reduce the run time of ABAP code (BADI) when it is reading huge data from BPC Cube and thus writing back huge data to the cube.

Hi All ,
In Case of reading huge amount of record from BPC Cube from BADI code , performing calculations and writing back huge amount of data into the cube , It takes lot of time . If there is any suggestion to read the data from Cube or writing data into the cube using some Parallel Processing methods , Then Please suggest .
Regards,
SHUBHAM

Hi Gersh ,
If we have a specific server say 10.10.10.10 (abc.co.in) on which we are working, Then under RZ12 we make the following entry as :
LOGON GROUP          INSTANCE
parallel_generators        abc.co.in_10         ( Lets assume : The instance number is 10 )
Now in SM59 under ABAP Connections , I am giving the following technical settings:
TARGET HOST          abc.co.in
IP address                  10.10.10.10
Instance number          10
Now if we have a scenario of load balancing servers with following server details (with all servers on different instance numbers ) :
10.10.10.11
10.10.10.13
10.1010.10
10.10.10.15
In this case how can we make the RZ12 settings and SM59 settings such that we don't have to hardcode any IP Address.
If the request is redirected to 10.10.10.11 and not to 10.10.10.10 , in that case how will the settings be.
I have raised this question on the below thread :
How to configure RZ12 and SM59 ABAP connection settings when we have work with Load Balancing servers rather than a specific server .
Regards,
SHUBHAM

How to reduce the open time of Ms proejct 2010 plans which are taking 6 minutes to open.

In our company there is many plans which are uploaded in Project server 2010.
all linking to each other and in master plan all also link with master project.
when we open any project plan in Ms Project 2010 then it is taking 5 to 7 minutes time.
I dont know why . it's too much time.

Is this issue specific to one master project? Do you have MS recommended db maintenance plan in place. Try updating database statics by executing SP_updatestats on all 4 project databases
Hrishi Deshpande Senior Consultant

How to reduce the query execution time

hai all,
We have created query on Purchasing Cube 0PUR_C01 for
                            Purchase Order (PO) analysis for single vendor materials, but it is taking long time to execute (about 45 sec...).
In the above Query we have used the following things:
In Columns:
i) Exceptional aggregation for maximum & minimum PO Net Price using reference characteristic as Calendar Day.
ii) Minimum PO Price value we have multiplied with Actual GR Quantity for the calculation of Impact of Lowest PO Net Price.
iii) Number of vendors calculated key figure.
In Rows:i)     Only Material
In Filters:
i)     Plant with variable select Option u2013 Optional.
ii)     Calendar Year / Month with Select Option u2013 Optional.
iii)     Material with excluded Unassigned (#).
iv)     Vendor with excluded Unassigned (#).
Following are we have used for Performance:
i)     Aggregates using Propose from query (only for this query).
ii)     Partitioning on Calendar Year / Month (For 1 year 14 partitions) i.e. (04.2007 to 03.2008).
iii)      Collapse.
iv)     In RSRT we have set the following properties
Read Mode = H
Req.Status = 0
Catch Mode = 4
Persistence Mode = 3 (BLOB)
Optimization mode = 0.
Our inputs to this Query:
i)     We are passing plant range 1201 to 1299.
ii)     Calendar Year / Month 04.2007 to 03.2008.
So please suggest me how to reduce the execution time.
please help me.
Thanks,
kiran manyam

Hi,
First of all its a complete question with all the details. Good work.
As you partitioned the cube based on calmonth and you are also giving calmonth in selection, it will definitely work towards improved query performance.
As you are putting plant values in the selection, is there any aggregate available on plant characteristics? If not creating a aggregate on plant will help.
Regards,
Yogesh

How to reduce the size of cloned Oracle Applications Instance

How to reduce the size of cloned Oracle applications instance, so as to save the storage space for the cloned systems.
How can we remove unimplemented modules in the instance so that we can reclaim the space occupied by them.
can any please suggest on this..?

mumbai,
I would recommend to leave it as it is.
If you can add some inexpensive HDD to your test system then do so.
The Apps DB storage decreasing process is quite painful and isn't straightforward.
You cant just delete unused schemas & tablesapces (if they are not used than they are not consuming reasonable space anyway). The modules in Apps are using each other procedures and objects. The effect of deleting one module (schema) is highly unpredictable.
If you will do something like that than you will not be able to patch your cloned system after that.
Some things can be done however in order to reduce size of a testing Apps system:
- Decries UNDO & TEMP tablespaces. Normally you do not need that big tbs in test system as you have got in the production.
- If your REDO LOG files are quite big and you have got a lot of redo groups you can recreated smaller REDO log files.
- On Apps tier you can delete OUTput and LOG files
In case you still would like to decrease data volume in your testing system you need to take a look on some tools which provide data subtracting capabilities from Apps DB. This process has to be quite intelligent. The tool have to know the data structures in APPS DB and subtract data in the way to not harm logical relationships between records. After a subtracting process you will need to make multiple FULL exp/imp cycles in order to reduce physical space consumed by the database. Beside the fact that those tools are quite expensive you will need to spend a lot of your time to implement those.
BTW: That is the size of you production DB? Have you analyzed which module takes most of the space? May be you can identify top 10 objects and try to archive a data from those object preventing the future grow of the production database.
Just my 0.02£
Yury
Check this out:
A.
http://www.freelists.org/archives/ora-apps-dba/05-2006/msg00000.html
B.
- Users can subscribe to your list by sending email to
ora-apps-dba-request_at_freelists.org with 'subscribe' in the Subject field
C.
http://www.freelists.org/archives/ora-apps-dba/05-2006/threads.html

Web report for opening it takes long time how to reduce the time?

HI Experts,
I created report using webi. For next time while opening it takes more time. How to reduce the time.

Hi Manikandan,
There could be multiple reason behind the bad performance.
1. Are you using supported JVM version to run Webi reports?
2. Do the reports contain prompts? Are the LOVs set to "Automatic refresh before use"? If yes, uncheck it.
3. How many queries do the report has? Can you run the queries one by one and check which query is taking much time?
4. If it based on relational database, copy and run it against the database directly and check how much time it takes. If more, you may have to fix the issue at database level like using temporary tables, indices, etc.
5. Also check if there are firewalls or proxy server between client and server communication. if yes, try to refresh the report directly on BO server and check.
6. Check if there is any hinderance between BO server and Database server communcation.
Hope it will help.
Regards,
Yuvraj

HT5731 How do I change my iTunes settings to lower the resolution of movies I download to reduce the download time?

How do I change my iTunes settings to lower the resolution of movies I download to reduce the download time?

If you're buying HD movies, you can change the setting in the Store preferences in iTunes to prefer 720p. Or you can just buy SD versions.
Regards.

How long does it take to connect to Wi-Fi the first time I use my Ipad? It's been over 30 minutes. I see that it wants to connect but it keeps on turning.

How long does it take to connect to Wi-Fi the first time I use my Ipad? It's been over 30 minutes. I see that it wants to connect but it keeps on turning.

It should not take very long. Up to thirty seconds tops depending on your Wi-Fi network.
Try rebooting your Wi-Fi router. Turn off the router for 30 seconds then turn it back on. After the router completes its startup routine and indicates a good connection to the internet start you iPad again and see if it will connect.
If you still have problems you might try connecting to a different Wi-Fi network. Maybe one at your local coffee shop or library.

How to restart the mac like the first time you use it?

how to restart the mac like the first time you use it?

Surefire way is by completely reinstalling the OS after erasing the Macintosh HD boot volume.
I think there's a way to coerce the Setup Assistant to run once again like on First Start, but requires erasing a couple of files in the drive. Don't know which ones right now.

How can I reduce the recording time of the webcam?

How can I reduce the recording time of the webcam?
thanks
Alex

Keep in mind that Logic also has a limit of 8550 quarter notes. So changing the time signature to 4/8 can get longer recording times. So make sure you change this setting along with the song end. See page 93 in the reference manual.
MBP 1.83 & MDD 1G DP Mac OS X (10.4.7) Logic Pro 7.2.2, Motu 828MkII+828, Firebox LC&XT, Tranzport, Unitor8+(3x)AMT-8's

How to reduce the cloning time if using cold backup?

Similar Messages

Maybe you are looking for