Need Clarification on sga_target and sga_max_size

HI,
I need some clarification in SGA_TARGET and SGA_MAX_SIZE.
I have the parameter like below.
SGA_MAX_SIZE=10G
SGA_TARGET=9G
And I spread the 9G to all components like (DB_CACHE,SHARE_POOL etc.,).
My doubt, Incase db need the memory more than 9GB, Whether it automatically take the extra 1G from sga_max_size
or we have to change the sga_target to 10G.

Unless and untill, we set the sga_taget=10G, The extra 1G(from sga_max_size) is not used.
Am i correct?No - its wrong. Any change in the value of SGA_TARGET affects only the sizes of the auto-tuned components. If you increases its value then increased memory will be distributed in only among the components controlled. So, yes 1GB will be used, because have sga_max_size=10G. If you decrease the value then reduced memory is taken back by the auto-tuning policy from one or more of the auto-tuned components.
If SGA_MAX_SIZE is greater than SGA_TARGET, you can increase SGA_TARGET without restarting the instance. Otherwise, you'd need to shutdown and restart the instance if you wanted to increase SGA_TARGET.
Regards
Girish Sharma

Similar Messages

  • Advice on sga_target and sga_max_size

    hi,
    a little abt my existing system
    HP UX 11.11 oracle 10.2.0.4 EBS11.5.10
    memory 4GB
    SGA 2 GB
    i will be increasing physical emory to 20GB today
    i need advice on setting sga_target and sga_max_size
    currently my SGA_MAX_SIZE=2GB

    If you are increasing the physical memory then set the SGA_MAX_SIZE to max estimated size based on application usage and keep the SGA_TARGET to whatever value you have currently.
    This helps to increase SGA dynamically (SGA_TARGET) when ever application needs and you no need to shutdown your database.
    It is always advised to keep SGA_MAX_SIZE > SGA_TARGET.
    Hope this helps.
    Regards.
    Satishbabu Gunukula
    http://oracleracexpert.blogspot.com
    [Click here to learn Install and Configure ASM in 10g|http://oracleracexpert.blogspot.com/2009/08/install-and-configure-asmlib-in-10g.html]
    Edited by: Satishbabu Gunukula on Aug 21, 2009 12:20 PM

  • Diffrence Between SGA_TARGET and SGA_MAX_SIZE

    Can Any one explain me what is the diference between
    SGA_TARGET and SGA_MAX_SIZE
    Thanks In Advance.

    sga_max_size -- This parameter sets the hard limit up to which sga_target can dynamically adjust sizes. At database start time, Oracle will allocate sga_max_size in RAM (or set sga_max_size to the sum of the existing pool sizes), so in order not to waste RAM it may be a good idea to have sga_max_size and sga_target at the same value, but there may be times when you want to have the capability to adjust for peak loads. By setting this parameter higher than sga_target, you allow dynamic adjustment of the sga_target parameter.
    sga_target -- This parameter is new in Oracle Database 10g and reflects the total size of memory footprint a SGA can consume. It includes in its boundaries the fixed SGA and other internal allocations, the (redo) log buffers, the shared pool, Java pool, streams pool, buffer cache, keep/recycle caches, and if they are specified, the non-standard block size caches.

  • WHAT WILL BE SGA_TARGET AND SGA_MAX_SIZE VALUE

    HI ALL
    RAM MEMORY 7360
    swap memory 8704 M
    OS--AIX6
    DATABASE SIZE ---50 GB
    WHAT WILL BE SGA_TARGET AND SGA_MAX_SIZE VALUE

    WHAT WILL BEThe values will be whatever you set them to be. OR sga_max_size will be automatically derived from db_cache_size + shared_pool_size + ............
    yes, I know that you are looking for a "thumb-rule" and your question is "_What *should* be ..."_ You might get many answers.
    The right answer is "what is really required based on the database, i/O pattern, concurrency etc
    Hemant K Chitale

  • Am I understanding SGA_TARGET and SGA_MAX_SIZE correctly?

    Let's say I set SGA_MAX_SIZE to 1024MB and SGA_TARGET to 512MB.
    As I understand it, this means that the total SGA size will never exceed 1024MB (that's the "hard cap"). Oracle will try to keep things around 512MB, but may burst up to 1024MB as needed.
    Is that correct? Otherwise, I don't see the point of SGA_MAX_SIZE vs. SGA_TARGET.
    BTW...I know MEMORY_TARGET and MEMORY_MAX_TARGET are the new 11g params, but Oracle ships XE with those turned off and the 10g SGA_* params and PGA_AGGREGATE_TARGET set...?

    Hi,
    In Oracle 11g MEMORY_TARGET is for automatic SGA and PGA management.MEMORY_MAX_SIZE is the maximum size to which your MEMORY_TARGET can be allocated.As shown below:
    SQL> show parameter %MEMORY%
    NAME TYPE VALUE
    hi_shared_memory_address integer 0
    memory_max_target big integer 1G
    memory_target big integer 1G
    But,When it comes to managing various other memory component like shared_pool,large_pool,Db_buffer_cache,log_buffer_cache,java_pool we still have to manage SGA_TARGET.
    SQL> show parameter %SGA%
    NAME TYPE VALUE
    lock_sga boolean FALSE
    pre_page_sga boolean FALSE
    sga_max_size big integer 1G
    sga_target big integer 0
    Best regards,
    Rafi.
    http://rafioracledba.blogspot.com/
    Edited by: Rafi (Oracle DBA) on Jun 22, 2011 11:54 PM

  • Need Clarification On Unicode and Upgrade-ECC6.0

    Dear All,
    I need some clarification unicode and upgrade . It would be great help if you give your time .
    We had 2 code pages - 1100,1401 in 4.6B system. We had languages - FR,EN,ES,PT and PL. The system has been upgraded to ECC6.0 non-unicode now.
    Now in I18N->System configuration (RSCPINST), only EN is listed. SPUMG asked for activation of I18N to proceed. When the I18N activation was done,  it has knocked out the code page 1401 from TCPDB table.
    Is this normal?
    But, the code pages 1401 is shown as consistent in SCP transaction.
    The system setting has changed to Single code page. Will this affect unicode migration? How did the additonal code page 1401 which was in 4.6B get knocked out
    now? How did the languages ES,FR,PT, IT and PL which were in 4.6B get knocked out of RCPINST?
    We are manually filling the vocabulary since SPUMG is not showing Scanning tabs. The language key in vocabulary is not completely set. The reprocess logs are not completely green. Will this allow unicode migration now? Can we start unicode migration even with this status.
    Regards,
    Santosh
    Edited by: santosh suryavanshi on Nov 18, 2010 11:11 PM
    Edited by: santosh suryavanshi on Nov 18, 2010 11:11 PM

    Hi Santosh,
    SAP ECC 6.0 is not supported with MDMP. This is the reason for the behaviour in RSCPINST.
    The standard way for an upgrade based on start release 4.6B with MDMP would be TU&UC (see SAP note 959698).
    Do you follow this procedure ?
    Best regards,
    Nils Buerckel
    SAP AG

  • SGA_TARGET and SGA_MAX_SIZE?

    Dear Professional,
    I have been asked by my client, He wants me to install Oracle 11g software in three servers. One server has  16GB another has 32 GB, and another has 64 GB of RAM? What would be the Best swap size? How memory I need to assign to SGA? what will be the SHmax and Shmin, kernal parameter? Please advice..
    Thanks
    Sagar

    The answer is 16GB, for all the machines.
    See here for reference:
    Oracle Database Quick Installation Guide
    Hope that helps,
    -Mark

  • Need clarification on SATA and PC3200

    First of all thanks to all who replied to my first question, I robbed a 3.5 drive out of an old computer and will use it to load driver then remove it.  Or I may copy those drivers to cd and see what happens. Is it possible to do this and not have to install a floppy at all? My main question is when I install the Sata hard drive on a brand new system will it work for me to install the OS (win xp) then install the drivers?  Whats with using either promise or intel SATA controller?  Also should I expect a problem using OCZ pc3200 dual channel ram on the first startup?  Sorry to be so dumb but I cant find this info on the net in any detail and I really need this to go smoothly my first time.  Here is a rundown on what I am putting together:
      Full tower ATX case, loaded with 5 fans
    420 watt PS with dual fans
    MSI neo FIS2R
    P4 3Ghz 800fsb
    MSI ti4800se 128mb ddr
    Seagate 80GB SATA
    Lite on 16x dvd rom
    logitech wireless optical moue and keyboard
    Win XP oem full version
    OCZ pc3200 1GB dual channel copper heat spreader
    ancient 3.5" floppy drive circa early 90's  (will be removed)
    Thank you again for your help.

    Copying the drivers on the Windows CD will not be suffient i'm afraid. My best guess is install the driver from an old floppy drive and afterwards you can remove it and disable the FDC controller in your BIOS.
    As for the dual channel DDR setup, i didn't experience any problems, so i don't think you will get any either.
    Last, the promise S-ATA/ATA controller is only for expansion. I've tested the speed of both controllers (ICH5R and the promise) and didn't find any shocking differences.

  • Needs Clarification Regarding Segments and Datafiles

    Hi,
    I want clarification regarding Segments, Datafiles and extents.
    As we know that A segment is made of one or more extents and extents are composed of one or more datablocks in the HD.
    Since all data are store in Datafiles which are composed of extents and datablocks. I want to know weather a table(Segment) can span to multiple datafiles or in a sigle datafile.
    Regards,
    D.Abbasi

    And an easy way to check it by yourself :
    SQL> create tablespace abbasi_tbs
      2  datafile 'E:\ORADATA\DEMO111P\abbasi_01.dbf' size 1m autoextend off,
      3           'E:\ORADATA\DEMO111P\abbasi_02.dbf' size 1m autoextend off;
    Tablespace created.
    SQL> create table abbasi_tbl (id number)
      2  tablespace abbasi_tbs;
    Table created.
    SQL> insert into abbasi_tbl
      2  select rownum as rn
      3  from   dual
      4  connect by level <=10000;
    10000 rows created.
    SQL> commit;
    Commit complete.
    SQL> select distinct file_id
      2  from   dba_extents
      3  where  segment_name ='ABBASI_TBL';
       FILE_ID
             6
             7
    or...
    SQL> select distinct DBMS_ROWID.ROWID_RELATIVE_FNO(rowid)
      2  from   abbasi_tbl;
    DBMS_ROWID.ROWID_RELATIVE_FNO(ROWID)
                                       6
                                       7
    SQL> select file_name from dba_data_files where file_id in (6,7);
    FILE_NAME
    E:\ORADATA\DEMO111P\ABBASI_01.DBF
    E:\ORADATA\DEMO111P\ABBASI_02.DBF
    SQL>Nicolas.
    added the ROWID function usage
    Edited by: N. Gasparotto on Jun 21, 2009 11:02 AM

  • Needed clarification on RWRKBENCH and CACHE

    Hi,
            I'm new to SAP-XI, Ihave some doubts regarding runtime work bench and cache refreshing
                   1.what is difference between complete cache refresh and delta cache refresh in SXI_CACHE transaction.If delta cache refresh is choosen then which components cache will be refreshed?
    2.In which case we have to go for complete cache or delta cache refresh ?
    3.what is difference between monitoring on RUNTIME WORK BENCH  & SXMB_MONI?
                   Plz if any body knows anwers to this questions kindly provide explanations.
    Good answers will be rewarded.
                                                                                    Regards,
                                                                                    Monica.

    Hi Monica,
    <b>1.what is difference between complete cache refresh and delta cache refresh in SXI_CACHE transaction.If delta cache refresh is choosen then which components cache will be refreshed?</b>
    1.Delta Cache Refresh: Any new objects that have been created/modified during the design/configuration activities are inserted into the temporary database tables when we use delta cache refresh option.
    2.Full Cache Refresh: All entries in temporary cache tables are deleted and a background program for inserting the complete information about all the design/configuration objects.
    <b>2.In which case we have to go for complete cache or delta cache refresh ?</b>
    check this
    /people/sravya.talanki2/blog/2005/12/02/sxicache--ripped-off
    How to handle cache
    https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/1a69ea11-0d01-0010-fa80-b47a79301290
    SXI_CACHE
    /people/sravya.talanki2/blog/2005/12/02/sxicache--ripped-off%3Fpage%3Dlast
    <b>3.what is difference between monitoring on RUNTIME WORK BENCH & SXMB_MONI?</b>
    In RWB we monitor messages whic goes via Adapter Engine.. i.e we monitor messages which r in JAVA stack
    In SXMB_MONI we monitor messages of Integartion Engine i.e messages of ABAP stack!!
    Hope u r now clear!!!
    regards
    biplab
    Use a Good Subject Line, One Question Per Posting - Award Points
    Message was edited by:
            biplab das

  • Sga_target and shared_pool_size

    Hi All,
    I am using 11GR2 on linux box. few days back , we hit the ora-4031 and we reported the issue to MOS, they then suggested to bump up the value of shared_pool_size to 3G, which is currently 1G.
    Here , we are using sga_target and sga_max_size to 30G and also when I checked the current shared pool size, it's shows around 6G.
    Since I am using the Automatic memory memory management along with the shared_pool_size parameter set to 1G. So, Do I really need to bump up the shared_pool_size as per MOS. doesn't this simply
    means that , irrespective of shared_pool_size being set , oracle will ignore this value and size the shared pool according to the load in the database.
    I would really appreciate , if you could offer your expert opinion here.
    SQL>  show parameter sga_
    NAME                                 TYPE                              VALUE
    sga_max_size                         big integer                       30G
    sga_target                           big integer                       30G
    SQL> show parameter shared
    NAME                                 TYPE                              VALUE
    hi_shared_memory_address             integer                           0
    max_shared_servers                   integer
    shared_memory_address                integer                           0
    shared_pool_reserved_size            big integer                       338899763
    shared_pool_size                     big integer                       1G
    shared_server_sessions               integer
    shared_servers                       integer                           0
    SQL> select sum(bytes)/1024/1024 mbytes
                  from v$sgastat where pool = 'shared pool';  2
                  MBYTES
                  6464
    Errors in file /db/prod01/oracle/admin/diag/rdbms/prod01/prod01/trace/prod01_m001_19998.trc:
    ORA-04031: unable to allocate 12312 bytes of shared memory ("shared pool","unknown object","KKSSP^281","kglseshtTable")
    Fri Aug 16 12:49:19 2013
    Process m001 died, see its trace file

    If you are on 11.2(as you have said) , I would recommend that you use AMM only. Also, unsetting the shared pool parameter completely would leave it on the mercy of the AMM and that may go wrong as well. Since you are already hitting 4031, probably the best bet would be to set up a fixed, large sized shared pool as suggested by MOS which, beyond its initial value, would be managed by AMM.
    Just my 2 cents
    Aman....

  • BPM, Workflow and Netweaver - Need Clarification

    Hi Guru,
    I am new to workflow, BPM and Netweaver.  I have several questions about those concepts.
    1. What is/are the different between workflow and webflow.  Which scenario should take workflow into consideration? Which scenario webflow can be applied to?
    2. What is/are the different between workflow and BPM.  If I am going to implement workflow in a company, do I need to implement business workflow as well as BPM?
    3. I need clarification on Netweaver platform and concept.
    4. What is/are the different between workflow in R/3 and workflow under Netweaver?
    5. If I am going to implement workflow integrated with R/3 and Outlook email, do I need to buy new wrokflow for Netweaver and Netweaver platform or alternatively, I can use business workflow module under R/3 system?
    Sorry for many questions asked. I am new to those products.  I am now working on software selection for workflow technology.  My company is going to implement new workflow to client.  Thank you very much.
    Cheers,

    Please ask only one (or closely related) questions per thread. This makes it a lot easier to get a good structure in the database of previously answered questions. While we are on the subject of previously answered questions, I think you should have a look at them....
    My suggestion is therefore:
    1. Close this thread
    2. Read the Frequently Answered Questions. Before you ask (here are many workflow answers)
    3. Search the forum.
    4. Create new threads if you have questions afterwards.

  • Domain Guideance and Clarification using SVN and an Export suggestion

    Hello Oracle SQL Data Modeler Support,
    Apologies if this has been documented somehwere and I have missed reading it, but have gone through the User Guide and cannot find the clarification I want regarding domains.
    1) WHAT IS BEST PRACTICE TO SAVE WHEN USING SVN
    From the forum I have picked up that the domains file is in the following directory:
    ~\datamodeler\datamodeler\types
    File name is 'defaultdomains.xml'
    When I come to save the file using SVN I get 'Choose versioned folder for storing system types'
    I assume this is where the domains file is stored.
    I require the Domains to be avialable centrally to all Designs I create, what should I do?
    a) Set the folder to ~\datamodeler\datamodeler\types
    b) Create a design called 'Domains' and store it in this folder
    c) Any thing you may suggest
    2) EXPORT OF DOMAIN FILE SUGGESTION
    This should be a quick win for you, can you please add an Export Domains function, seems this needs to do no more than make a copy of the defaultdomains.xml file and create it in a specified export directory.
    Will avoid having to go through the forum to pick up that the defaultdomains.xml file needs to be copied and transfered over for new SQL Data Modeler installations.

    Hello,
    I require the Domains to be avialable centrally to all Designs I create, what should I do?Default location is fine if SVN is not used and if all designs are used only on that computer.
    If versioning is used then it's better to have separate directory for domains and this directory shouldn't be part of any design's directory - i.e. for designs you can have directories c:\des_1, c:\des_2 ...c:\des_n - one directory per each design and that directory will contain design DMD file and design folder. For domains you can have directory c:\DM_Sys_types and you need to set this directory in "Tools>Preferences>Data Modeler>system types directory" - logical types, RDBMS sites and scripts also will be stored there.
    Philip

  • Need clarification on issue with tablespaces

    Hi All,
    I needed clarification on this issue with tablespaces.I am following a guideline which says "Presently,The general rule of thumb is to concentrate on tablespaces that are 90% or greater used and have less than 2 GB of free space"
    My Scenario:-Now i have tablesace which is 97.23 % and the free space is 10 gb.Please advise me that whats the best next step I can take in this respect?
    I am very new to this and just following the guideline,so can you all please explain me some details about this.
    Thanks

    Well, the guideline says "*and* have less than 2 GB of free space"
    Since that tablespace as 10GB of free space, the guideline says that you do not need to concentrate on it. Eh ?
    I would be careful with guidelines that are hard-coded in this manner. You need to know the context in which the guideline was set.
    Assuming that your database/application doesn't grow suddenly by 1GB or larger (or create temporary segments / objects of 1GB or larger), the "2 GB of free space" might make sense. Then, again, it depends on the extent allocation type. What size are extents getting allocated.
    You should check with your organisation's senior DBAs on what happens if you don't concentrate on a tablespace because it is outside the guidelines !

  • Need clarification on Bigfile Tablespaces

    In the following Oracle Documentation Lirbary PDF,
        Oracle® Database
        Concepts
        10g Release 2 (10.2)
        B14220-02
        October 2005
    section
        Overview of Tablespaces
        Bigfile Tablespaces (page: 3-5)
    it says,
        Benefits of Bigfile Tablespaces
        * Bigfile tablespaces can significantly increase the storage capacity of an Oracle database. Smallfile tablespaces can contain up to 1024 files, but bigfile tablespaces contain only one file that can be 1024 times larger than a smallfile tablespace. The total tablespace capacity is the same for smallfile tablespaces and bigfile tablespaces. However, because there is a limit of 64K database for each database, a database can contain 1024 times more bigfile tablespaces than smallfile tablespaces, so bigfile tablespaces increase the total database capacity by 3 orders of magnitude. In other words, 8 exabytes is the maximum size of the Oracle database when bigfile tablespaces are used with the maximum block size (32k).
    I need clarification on how to arrive at 8 exabytes ?
    1024 x 32k x 64,000 ??
    According to the exerpt above, there's no mention of maximum number of Operating System blocks per extent. Unless this was assumed knowledge ... how do I get 8 exabytes ?
    And if "a database can contain 1024 times more bigfile tablespaces than smallfile tablespaces", then what's the upper limit on smallfile tablespaces ? -- was this sentence referring to the number of datafiles per smallfile tablespace ? ...
    O_o
    Thanks !
    Message was edited by:
    mvanle

    Hi,
    According to [url http://download-west.oracle.com/docs/cd/B19306_01/server.102/b14237/limits002.htm#i287915]Physical Database Limits page, a bigfile tablespace contains only one datafile or tempfile, which can contain up to approximately 4 billion ( 232 ) blocks. The maximum size of the single datafile or tempfile is 128 terabytes (TB) for a tablespace with 32K blocks and 32TB for a tablespace with 8K blocks. In resume, A bigfile tablespace is a tablespace containing a single datafile that can be as large as 128 terabytes (TB), depending on the block size of the tablespace. In conjunction with setting the initialization parameter DB_FILES to the maximum value of 65,635 the total size of the database can be more than 8 exabytes (EB)
    >>how do I get 8 exabytes ?
    You can calculate the maximum amount of space (M) in a single Oracle database as the maximum number of datafiles (D) multiplied by the maximum number of blocks per datafile (F) multiplied by the tablespace block size (B):
    M = D * F * B. Therefore, the maximum database size, given the maximum block size and the maximum number of datafiles, is:
    65,535 datafiles * 4,294,967,296 blocks per datafile * 32,768 block size = 9,223,231,299,366,420,480 = 8EB.Cheers
    Legatti

Maybe you are looking for