Lob Storage

In "Oracle Database Application Developer's Guide - Large Objects
10g Release 2 (10.2)", Section "Inline and Out-of-Line LOB Storage", it says:
- "LOB values are stored out-of-line when any of the following situations apply: By default. That is, if you do not specify a LOB parameter for the LOB storage clause when you create the table."
Some lines down it says
"Using the default LOB storage properties (inline storage) ".
Does any one known what is the default option?
By outline it means that the LOB data is not in the same page as the other fields data, but it is still in the same tablespace. Isn't that?
Thanks

it means that the LOB data is not in the same page as the other fields data,
but it is still in the same tablespace. Isn't that?Nothing better than trying...
TEST@db102 SQL> create table test_lob (a number, b clob);
Table created.
TEST@db102 SQL> select segment_name,segment_type,tablespace_name from user_segments
  2  where segment_name = 'TEST_LOB'
  3  union
  4  select segment_name,segment_type,tablespace_name from user_segments
  5  where segment_name in (select segment_name from user_lobs
  6  where table_name = 'TEST_LOB');
SEGMENT_NAME                             SEGMENT_TYPE       TABLESPACE_NAME
SYS_LOB0000053644C00002$$                LOBSEGMENT         USERS
TEST_LOB                                 TABLE              USERS
TEST@db102 SQL>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               

Similar Messages

  • LOB Storage in Designer 9i

    Hi,
    In the help for Designer 9i I found a note
    explaining the different parameters which can be set in a LOB storage clause.
    Does anybody know where these parameters can be set for a given table implementation?
    Regards Bernd

    Now, I have found that "the repository reports" tool has predefined documentation reports. When I request a document, the operation never ends. When I kill the tool a word document appears with the information. I don't know if it is a problem of the current design (not complete, incorrect in some way) or if it is a problem of my pc (RAM, Virtual Memory). I can almost assure that it is not the second case but I don't know how to avoid the problem of this hanging of the tool.

  • LOB Storage definition in Designer

    Does somebody know how to specify a
    LOB Storage definition in Designer 6i?
    Thanks,
    Michael

    Hi Didier - its actually the measure I'm trying tu restrict.
    I'm a bit hazy on what I'øm supposed to use as a filter key.
    What I'm trying to do is:
    1. Create a restricted measuer called BO Amount Curr
    2. Which is based on SAP measure Amount
    3. The restriction should be based on the dimension L01 Currency
    4. where the use selects the desired currency unit at run time.
    I suspect that I'm messing up the filter key selection - however, I haven't been able to find a guide on how to use this functionality.
    What should be uesd as Filter key - the dimesion you want to filter or the dimension you want to use as filter?
    When I use the Filter dimension as filter key - I get an empty query result.
    br
    Jess

  • Buffer busy waits after cnanging lob storage  to oracle securefiles

    Hi Everyone
    I need help resolving a problem with buffer busy waits in for a lob segment using securefiles for storage.
    During the load the application inserts a record into a table with the lob segment and update the record after, populating lob data. The block size on the table space holding the lob is 8 kb and the chunk size on the lob segment is set to 8kb. The average size of the lob record is 6 kb and the minimum size is 4.03 KB. The problem occurs only when running a job with a big number of relatively small inserts (4.03 Kb) in to the lob column . The table definition allow in-row storage and the ptcfree set to 10%. The same jobs runs without problem when using basicfiles storage for the lob column.
    According to [oracle white paper |http://www.oracle.com/technetwork/database/options/compression/overview/securefiles-131281.pdf] securefiles have a number of performance enhancements. I was particular interested to test Write Gather Cache as our application does a lot of relatively small inserts into a lob segment.
    Below is a fragment from the AWR report. It looks like all buffer busy waits belong to a free list class. The lob segment is located in an ASSM tablespace and I cannot increase freelists.
    Connected to:
    Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - 64bit Production
    With the Partitioning option
    Host Name        Platform                         CPUs Cores Sockets Memory(GB)
    DB5              Microsoft Windows x86 64-bit        8     2              31.99
                  Snap Id      Snap Time      Sessions Curs/Sess
    Begin Snap:      1259 01-Apr-11 14:40:45       135       5.5
      End Snap:      1260 01-Apr-11 15:08:59       155      12.0
       Elapsed:               28.25 (mins)
       DB Time:              281.55 (mins)
    Cache Sizes                       Begin        End
    ~~~~~~~~~~~                  ---------- ----------
                   Buffer Cache:     2,496M     2,832M  Std Block Size:         8K
               Shared Pool Size:     1,488M     1,488M      Log Buffer:    11,888K
    Load Profile              Per Second    Per Transaction   Per Exec   Per Call
    ~~~~~~~~~~~~         ---------------    --------------- ---------- ----------
          DB Time(s):               10.0                0.1       0.01       0.00
           DB CPU(s):                2.8                0.0       0.00       0.00
           Redo size:        1,429,862.3            9,390.5
       Logical reads:          472,459.0            3,102.8
       Block changes:            9,849.7               64.7
      Physical reads:               61.1                0.4
    Physical writes:               98.6                0.7
          User calls:            2,718.8               17.9
              Parses:              669.8                4.4
         Hard parses:                2.2                0.0
    W/A MB processed:                1.1                0.0
              Logons:                0.1                0.0
            Executes:            1,461.0                9.6
           Rollbacks:                0.0                0.0
        Transactions:              152.3
    Top 5 Timed Foreground Events
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
                                                               Avg
                                                              wait   % DB
    Event                                 Waits     Time(s)   (ms)   time Wait Class
    buffer busy waits                 1,002,549       8,951      9   53.0 Concurrenc
    DB CPU                                            4,724          28.0
    latch: cache buffers chains      11,927,297       1,396      0    8.3 Concurrenc
    direct path read                    121,767         863      7    5.1 User I/O
    enq: DW - contention                209,278         627      3    3.7 Other
    ?Host CPU (CPUs:    8 Cores:    2 Sockets: )
    ~~~~~~~~         Load Average
                   Begin       End     %User   %System      %WIO     %Idle
           38.7       3.5       57.9
    Instance CPU
    ~~~~~~~~~~~~
                  % of total CPU for Instance:      40.1
                  % of busy  CPU for Instance:      95.2
      %DB time waiting for CPU - Resource Mgr:       0.0
    Memory Statistics
    ~~~~~~~~~~~~~~~~~                       Begin          End
                      Host Mem (MB):     32,762.6     32,762.6
                       SGA use (MB):      4,656.0      4,992.0
                       PGA use (MB):        318.4        413.5
        % Host Mem used for SGA+PGA:        15.18        16.50
                                                                 Avg
                                            %Time Total Wait    wait    Waits   % DB
    Event                             Waits -outs   Time (s)    (ms)     /txn   time
    buffer busy waits             1,002,549     0      8,951       9      3.9   53.0
    latch: cache buffers chain   11,927,297     0      1,396       0     46.2    8.3
    direct path read                121,767     0        863       7      0.5    5.1
    enq: DW - contention            209,278     0        627       3      0.8    3.7
    log file sync                   288,785     0        118       0      1.1     .7
    SQL*Net more data from cli    1,176,770     0        103       0      4.6     .6
    Buffer Wait Statistics                DB/Inst: ORA11G/ora11g  Snaps: 1259-1260
    -> ordered by wait time desc, waits desc
    Class                    Waits Total Wait Time (s)  Avg Time (ms)
    free list              818,606               8,780             11
    undo header            512,358                 141              0
    2nd level bmb          105,816                  29              0
    -> Total Logical Reads:     800,688,490
    -> Captured Segments account for   19.8% of Total
               Tablespace                      Subobject  Obj.       Logical
    Owner         Name    Object Name            Name     Type         Reads  %Total
    EAG50NSJ   EAG50NSJ   SYS_LOB0000082335C00            LOB    127,182,208   15.88
    SYS        SYSTEM     TS$                             TABLE    7,641,808     .95
    Segments by Physical Reads            DB/Inst: ORA11G/ora11g  Snaps: 1259-1260
    -> Total Physical Reads:         103,481
    -> Captured Segments account for  224.4% of Total
               Tablespace                      Subobject  Obj.      Physical
    Owner         Name    Object Name            Name     Type         Reads  %Total
    EAG50NSJ   EAG50NSJ   SYS_LOB0000082335C00            LOB        218,858  211.50
    ....Best regards
    Yuri Kogun

    Hi Jonathan,
    I was puzzled by the number of logical reads as well. This hasn't happened when the lob was stored as a basic fille and I assumed that the database is able to store the records in-row when we switched to securefiles. With regards to ASSM, according to the documentation this is the only option when using securefiles.
    We did have high number of HW-enqueue waits in the database when running the test with basic files and had to set 44951 event
    alter system set EVENTS '44951 TRACE NAME CONTEXT FOREVER, LEVEL 1024' There are 2 application servers running 16 jobs each, so we should not have more than 32 sessions inserting the data in the same time but I need to check wheter jobs can be brocken to smaller peaces. I that case the number of concurrent session may be bigger. Each session is configured with bundle size of 30 and it will issue commit every 30 inserts.
    I am not sure how exactly the code does insert, as I've been told it should be straight insert and update I will be able to check this on Monday.
    Below is the extract from the AWR reports with top SQL, I could not find any SQL related to the $TS table in the report. The query to the V$SEGMENT_STATISTICS was executed by me during the job run.
    ?SQL ordered by Elapsed Time          DB/Inst: ORA11G/ora11g  Snaps: 1259-1260
    -> Resources reported for PL/SQL code includes the resources used by all SQL
       statements called by the code.
    -> % Total DB Time is the Elapsed Time of the SQL statement divided
       into the Total Database Time multiplied by 100
    -> %Total - Elapsed Time  as a percentage of Total DB time
    -> %CPU   - CPU Time      as a percentage of Elapsed Time
    -> %IO    - User I/O Time as a percentage of Elapsed Time
    -> Captured SQL account for   91.3% of Total DB Time (s):          16,893
    -> Captured PL/SQL account for    0.1% of Total DB Time (s):          16,893
            Elapsed                  Elapsed Time
            Time (s)    Executions  per Exec (s)  %Total   %CPU    %IO    SQL Id
             7,837.5        119,351          0.07   46.4   28.3     .7 2zrh6mw372asz
    Module: JDBC Thin Client
    update JS_CHANNELDESTS set CHANNELID=:1, DESTID=:2, CHANNELDESTSTATUSDATE=:3, ST
    ATUS=:4, BINOFFSET=:5, BINNAME=:6, PAGECOUNT=:7, DATA=:8, SORTORDER=:9, PRINTFOR
    MAT=:10, ENVELOPEID=:11, DOCID=:12, CEENVELOPEID=:13, CHANNELTYPE=:14 where ID=:
    15
             7,119.0        115,997          0.06   42.1   23.1     .2 3vjx93vur4dw1
    Module: JDBC Thin Client
    insert into JS_CHANNELDESTS (CHANNELID, DESTID, CHANNELDESTSTATUSDATE, STATUS, B
    INOFFSET, BINNAME, PAGECOUNT, DATA, SORTORDER, PRINTFORMAT, ENVELOPEID, DOCID, C
    EENVELOPEID, CHANNELTYPE, ID) values (:1, :2, :3, :4, :5, :6, :7, :8, :9, :10, :
    11, :12, :13, :14, :15)
                85.6              2         42.80     .5   98.3     .0 cc19qha9pxsa4
    Module: SQL Developer
    select object_name, statistic_name, value from V$SEGMENT_STATISTICS
    where object_name = 'SYS_LOB0000082335C00011$$'
                35.0        111,900          0.00     .2   74.3    7.6 c5q15mpnbc43w
    Module: JDBC Thin Client
    insert into JS_ENVELOPES (BATCHID, TRANSACTIONNO, SPOOLID, JOBSETUPID, JOBSETUPN
    AME, SPOOLNAME, STEPNO, MASTERCHANNELJOBID, SORTKEY1, SORTKEY2, SORTKEY3, ID) va
    lues (:1, :2, :3, :4, :5, :6, :7, :8, :9, :10, :11, :12)
                34.9        111,902          0.00     .2   63.0    2.6 a0hmmbjwgwh1k
    Module: JDBC Thin Client
    insert into JS_CHANNELJOBPROPERTIES (NAME, VALUE, CHANNELJOBID, ID) values (:1,
    :2, :3, :4)
                29.2            950          0.03     .2   95.9     .1 du0hgjbn9vw0v
    Module: JDBC Thin Client
    SELECT * FROM JS_BATCHOVERVIEW WHERE BATCHID = :1
    ?SQL ordered by Executions            DB/Inst: ORA11G/ora11g  Snaps: 1259-1260
    -> %CPU   - CPU Time      as a percentage of Elapsed Time
    -> %IO    - User I/O Time as a percentage of Elapsed Time
    -> Total Executions:       2,476,038
    -> Captured SQL account for   96.0% of Total
                                                  Elapsed
    Executions   Rows Processed  Rows per Exec   Time (s)   %CPU    %IO    SQL Id
         223,581         223,540            1.0       22.4   63.7     .0 gz7n75pf57c
    Module: JDBC Thin Client
    SELECT SQ_CHANNELJOBPROPERTIES.NEXTVAL FROM DUAL
         120,624         120,616            1.0        8.1   99.0     .0 6y3ayqzubcb
    Module: JDBC Thin Client
    select batch0_.BATCHID as BATCHID0_0_, batch0_.BATCHNAME as BATCHNAME0_0_, batch
    0_.STARTDATE as STARTDATE0_0_, batch0_.PARFINDATE as PARFINDATE0_0_, batch0_.PRO
    CCOMPDATE as PROCCOMP5_0_0_, batch0_.BATCHSTATUS as BATCHSTA6_0_0_, batch0_.DATA
    FILE as DATAFILE0_0_, batch0_.BATCHCFG as BATCHCFG0_0_, batch0_.FINDATE as FINDA
         119,351         227,878            1.9    7,837.5   28.3     .7 2zrh6mw372a
    Module: JDBC Thin Client
    update JS_CHANNELDESTS set CHANNELID=:1, DESTID=:2, CHANNELDESTSTATUSDATE=:3, ST
    ATUS=:4, BINOFFSET=:5, BINNAME=:6, PAGECOUNT=:7, DATA=:8, SORTORDER=:9, PRINTFOR
    MAT=:10, ENVELOPEID=:11, DOCID=:12, CEENVELOPEID=:13, CHANNELTYPE=:14 where ID=:
    15
         116,033         223,892            1.9        8.0   92.2     .0 406wh6gd9nk
    Module: JDBC Thin Client
    select m_jobprope0_.CHANNELJOBID as CHANNELJ4_1_, m_jobprope0_.ID as ID1_, m_job
    prope0_.NAME as formula0_1_, m_jobprope0_.ID as ID4_0_, m_jobprope0_.NAME as NAM
    E4_0_, m_jobprope0_.VALUE as VALUE4_0_, m_jobprope0_.CHANNELJOBID as CHANNELJ4_4
    _0_ from JS_CHANNELJOBPROPERTIES m_jobprope0_ where m_jobprope0_.CHANNELJOBID=:1
         115,997         115,996            1.0    7,119.0   23.1     .2 3vjx93vur4d
    Module: JDBC Thin Client
    insert into JS_CHANNELDESTS (CHANNELID, DESTID, CHANNELDESTSTATUSDATE, STATUS, B
    INOFFSET, BINNAME, PAGECOUNT, DATA, SORTORDER, PRINTFORMAT, ENVELOPEID, DOCID, C
    EENVELOPEID, CHANNELTYPE, ID) values (:1, :2, :3, :4, :5, :6, :7, :8, :9, :10, :
    11, :12, :13, :14, :15)
         115,996         115,996            1.0       15.9   75.0    4.5 3h58syyk145
    Module: JDBC Thin Client
    insert into JS_DOCJOBS (CREATEDATE, EFFDATE, JURIST, LANG, IDIOM, DD, DDVID, USE
    RKEY1, USERKEY2, USERKEY3, USERKEY4, USERKEY5, USERKEY6, USERKEY7, USERKEY8, USE
    RKEY9, USERKEY10, USERKEY11, USERKEY12, USERKEY13, USERKEY14, USERKEY15, USERKEY
    16, USERKEY17, USERKEY18, USERKEY19, USERKEY20, REVIEWCASEID, ID) values (:1, :2
         115,440         115,422            1.0       11.5   63.3     .0 2vn581q83s6
    Module: JDBC Thin Client
    SELECT SQ_CHANNELDESTS.NEXTVAL FROM DUAL
    ...The tablespace holding the lob segment has system extent allocation and the number of blocks for the lob segments roughly the same as the number of blocks in allocated extents.
           select segment_name,  blocks, count (*)  
           from dba_extents where segment_name = 'SYS_LOB0000082335C00011$$'
           group by segment_name,  blocks
           order by blocks
    SEGMENT_NAME                                                                      BLOCKS                 COUNT(*)              
    SYS_LOB0000082335C00011$$                                                         8                      1                     
    SYS_LOB0000082335C00011$$                                                         16                     1                     
    SYS_LOB0000082335C00011$$                                                         128                    158                   
    SYS_LOB0000082335C00011$$                                                         256                    1                     
    SYS_LOB0000082335C00011$$                                                         1024                   120                   
    SYS_LOB0000082335C00011$$                                                         2688                   1                     
    SYS_LOB0000082335C00011$$                                                         8192                   117                   
    SELECT
        sum(ceil(dbms_lob.getlength(data)/8000))
        from EAG50NSJ.JS_CHANNELDESTS
    SUM(CEIL(DBMS_LOB.GETLENGTH(DATA)/8000))
    993216                                  
    select sum  (blocks)   from dba_extents where segment_name = 'SYS_LOB0000082335C00011$$'
    SUM(BLOCKS)           
    1104536                Below is the instance activity stats related to securefiles from the AWR report
    Statistic                                     Total     per Second     per Trans
    securefile allocation bytes           3,719,995,392    2,195,042.4      14,415.7
    securefile allocation chunks                380,299          224.4           1.5
    securefile bytes non-transformed      2,270,735,265    1,339,883.4       8,799.6
    securefile direct read bytes          1,274,585,088      752,089.2       4,939.3
    securefile direct read ops                  119,725           70.7           0.5
    securefile direct write bytes         3,719,995,392    2,195,042.4      14,415.7
    securefile direct write ops                 380,269          224.4           1.5
    securefile number of non-transfo            343,918          202.9           1.3Best regards
    Yuri
    Edited by: ykogun on 02-Apr-2011 13:33

  • Retrieving LOB Storage Parameter through jdbc

    Is there a way to retrieve the lob column's initial and next storage parameter through jdbc?
    I need to replicate a table in one database to another database and with tables containing lobs(clobs,blobs),
    I need to get the storage parameter values for creating a "Create Table" script on the other database.
    tnx

    In records.jsp change your line that reads:
    out.print("<a href =vdetails.jsp>" +  plantno);Change it to:
    out.print("<a href =vdetails.jsp?plantno=" + plantno + ">" +  plantno);Then you can change your line in vdetails.jsp:
    String pla=request.getParameter("something");to
    String pla=request.getParameter("plantno");That should be enough to get you started.
    You should investigate either the use of a bean or even a custom tag to remove these scriplets from your JSP.

  • LOB Storage Attribute Changed Unexpectedly

    Hello All,
    I have an Oracle Text application and after testing high volume inserts (100,000 Docs) I noticed the LOB text column's IN_ROW storage attribute changed from out of line to in line storage. This had a negative effect on text search performance and I am positive the table was created and data initially loaded with out of line storage on my LOB.
    Does anyone know of any Oracle conditions or situations that would cause this to occur?
    Thanks..
    Joe

    Please change nologging in my earlier response to logging.
    alter table parttest
    modify partition pt_violation_200201
    lob(explanation) (nocache logging)

  • Questions on Lob-Columns

    Hi,
    i´m working still along time with Oracle, but not very often with Lob-Types.
    The docs and metalink didn´t helped me very much with the following questions:
    1. What is the maximum length of a CLOB Datatype in PL/SQL (long ago it was 32767 Bytes, but now (10.2) ?)
    2. Is empty_blob()/empty_clob() still necessary in 10.2? or when do i use it?
    3. What is the maximum length of a pl/sql clob datatype when i update a clob column in a table?
    declare
    c clob;
    begin
    c:=.......... very long;
    update t set col=c where id=1;
    end;
    4. Does a select in pl/sql return a locator or the data ob the lob (If locator, is there a easy way the read the contents of a clob without dbms_lob.read?)
    5. Is there dbms_lob Package faster/better/??? then using insert/update/delete direct on the column?
    Thanks
    Marco
    Message was edited by:
    mpatzwah

    You may find the following link useful:
    http://download.oracle.com/docs/cd/B19306_01/server.102/b14200/sql_elements001.htm#sthref173
    E.g., search for CLOB yields:
    CLOB objects can store up to (4 gigabytes -1) * (the value of the
    CHUNK parameter of LOB storage) of character data.
    If the tablespaces in your database are of standard block size,
    and if you have used the default value of the CHUNK parameter
    of LOB storage when creating a LOB column, then this is
    equivalent to (4 gigabytes - 1) * (database block size).
    CLOB objects have full transactional support. Changes made
    through SQL, the DBMS_LOB package, or the Oracle Call
    Interface (OCI) participate fully in the transaction.
    CLOB value manipulations can be committed and rolled back.
    However, you cannot save a CLOB locator in a PL/SQL or
    OCI variable in one transaction and then use it in another
    transaction or session.

  • How many LOBs in a table?

    Does it make sense having a table with 13 CLOBs? It seems to me that it would be better having not more than one LOB in a table. Do I remember well? In the DB I've been asked to tune and optimize, I've found such situations. Is it terrible or am I mistaken? Thanks!

    MichaelS wrote:
    LOB's are stored in seperate space outside the table with a LOB pointer stored on the tableNot generally: Inline and Out-of-Line LOB Storage, e.g. the smaller sized lobs can be stored inline.Ooo, yeah, I forgot about the inline ones. Good reminder thanks.

  • Oracle RAW / Long / LOBs - Buffer cache?

    Hello guys,
    i think i have read sometime ago something about that LOBs are not cached in the buffer cache? Is this right?
    I also think that i can remember that raw or long are stored in the sga buffer cache?
    I can not find any official documentation about that topic... maybe you can help me...
    Thanks and Regards
    Stefan

    Hi,
    Depends on whether you are on about temporary lobs or lobs that are persisted as a column. If it's as a column then you have the cache/cache read/nocache option which is part of the lob storage clause. If it's nocache it's not put into the buffer cache and if it's cache it is (and I'm sure you can figure out from this what "cache read" means)
    If it's a temporary lob then it exists in the temporary tablespace which if a true temp tablespace (which it should be) will be accessed using direct IO (so not buffered).
    An the official documentation is easily found, Oracle wrote a whole manual dedicated to LOBS...
    http://download.oracle.com/docs/cd/B19306_01/appdev.102/b14249/toc.htm
    HTH
    Chris

  • [SOLVED] Create table and storage clause

    Hi, I create a table whit lob ad so, I set some LOB storage clause:
    CREATE TABLE W2_T_MESSAGE_NWSL (
    KEY NUMBER,
    T_MESSAGE CLOB DEFAULT EMPTY_CLOB(),
    T_TEXTMESSAGE CLOB DEFAULT EMPTY_CLOB(),
    T_FOOTER CLOB DEFAULT EMPTY_CLOB())
    LOB (T_MESSAGE)
    STORE AS (TABLESPACE LOBDATA CACHE READS RETENTION ENABLE STORAGE IN ROW),
    LOB (T_TEXTMESSAGE)
    STORE AS (TABLESPACE LOBDATA CACHE READS RETENTION ENABLE STORAGE IN ROW),
    LOB (T_FOOTER)
    STORE AS (TABLESPACE LOBDATA CACHE READS RETENTION ENABLE STORAGE IN ROW);I need to export this table and do an import on a different instance.
    I know for sure that the tablespace LOBDATA exists on both database but I can't say the same for the user default tablespace.
    The problem is that when I import this table, the statement fail because the default tablespace of the user hasn't the same name of the tablespace where the table was exported.
    IMP-00003: ORACLE error 959 encountered
    ORA-00959: tablespace 'USERWNSTARTUP' does not exist
    This because the create table generated by the export has this storage clause:
    PCTFREE 10 PCTUSED 40 INITRANS 1 MAXTRANS 255 STORAGE(
    INITIAL 65536 FREELISTS 1 FREELIST GROUPS 1) TABLESPACE USERWNSTARTUP LOGGING NOCOMPRESS LOB (B_LOGO) STORE AS (TABLESPACE "LOBDATA" ENABLE STORAGE IN ROW CHUNK 8192 PCTVERSION 10 CACHE READS STORAGE(INITIAL 65536 FREELISTS 1 FREELIST GROUPS 1)
    On other tables, where I havent defined any storage clause all works fine.
    Is there a way to resolve this problem?
    Thanks
    Regards

    if someone is interested I've found this on the official import oracle documentation:
    if the tablespace no longer exists or the user does not have the necessary quota, the system uses the default tablespace for that user as long as the table is unpartitioned, contains no LOB or VARRAY columns, is not a type table, and is not an index-only table with an overflow segment.In my case I've a LOB column, so tablespace information would be exporterd. I've done a trace of the import and I can confirm that on table without lob, varray etc.. the create table statement hasn't the tablespace specification.

  • How to export and import LOB

    In my user_object, I have an object_type of LOB. How do I export and import object of this type.

    If you are on 10g try use data pump then you don't need to worry about anything.
    If you can only use exp/imp.
    You can export as usual, when import
    either you have the tablespace with exact name precreated for LOB storage
    or if you want to change LOB storage tablespace, you can precreate the tables having LOB type.

  • LOB and retention

    Hi,
    I try to modify the retention parameter on a partition table. Whatever the SQL order, I ran ... I can't see the result in the dictionary. What's wrong in my sql order ?
    Thanks a lot for your help ..
    The release of the database is 9.2.0.8.
    I create the table:
    ==============
    CREATE TABLE test_lob (chp1 clob, chp2 int)
    lob (chp1) STORE AS SEGNAME (TABLESPACE FD
    PCTVERSION 0)
    partition by range (chp2)
    (PARTITION P1 VALUES LESS THAN (1),
    PARTITION P2 VALUES LESS THAN (2),
    PARTITION P3 VALUES LESS THAN (3));
    I verify that the value is correct in the dictionary:
    ===================================
    select bitand(defflags,32) from sys.partlob$ where TABOBJ# in(select OBJECT_ID from dba_objects where OBJECT_NAME='TEST_LOB');
    (result is 0)
    I modify the table:
    ===============
    alter table test_lob modify lob(chp1) (retention);
    I check one more time:
    ==================
    select bitand(defflags,32) from sys.partlob$ where TABOBJ# in(select OBJECT_ID from dba_objects where OBJECT_NAME='TEST_LOB');
    (result is 0 .. must be 32)
    I make another SQL order:
    ====================
    alter table test_lob modify partition p1 lob (chp1) (retention);
    alter table test_lob modify partition p2 lob (chp1) (retention);
    alter table test_lob modify partition p3 lob (chp1) (retention);
    I check one more time :
    ====================
    select bitand(defflags,32) from sys.partlob$ where TABOBJ# in(select OBJECT_ID from dba_objects where OBJECT_NAME='TEST_LOB');
    (result is 0 .. must be 32)
    What's wrong ???

    Before I checked the 10g manual I would have said that LOB segments do not use UNDO tablespace. UNDO for LOB segments is kept in the LOB segment.
    >>
    PCTVERSION integer Specify the maximum percentage of overall LOB storage space used for maintaining old versions of the LOB. The default value is 10, meaning that older versions of the LOB data are not overwritten until they consume 10% of the overall LOB storage space.
    You can specify the PCTVERSION parameter whether the database is running in manual or automatic undo mode. PCTVERSION is the default in manual undo mode. RETENTION is the default in automatic undo mode. You cannot specify both PCTVERSION and RETENTION.
    RETENTION Use this clause to indicate that Oracle Database should retain old versions of this LOB column. Oracle Database uses the value of the UNDO_RETENTION initialization parameter to determine the amount of committed undo data to retain in the database.
    You can specify the RETENTION parameter only if the database is running in automatic undo mode. In this mode, RETENTION is the default value unless you specify PCTVERSION. You cannot specify both PCTVERSION and RETENTION
    <<
    I still believe that when RETENTION is set that Oracle uses the LOB segment to store the UNDO rather than the UNDO tablespace, but I will have to see the documentation in more detail to try to find support.
    HTH -- Mark D Powell --

  • Oracle 9.2 Compress LOB SEGMENT

    Hi
    I have a table with a BLOB column
    create table t (..., blob_column BLOB, ...) LOB (blob_column) STORE AS (tablespace lobts ...)
    In Oracle 11g it is possible to use COMPRESS [HIGH | MEDIUM] option for LOB storage
    I am using Oracle 9.2 and I would like to keep the BLOB data compressed in the tablespace lobts
    What if set tablespace lobts to have DEFAULT COMPRESS option:
    ALTER TABLESPACE lobts DEFAULT COMPRESS;
    Does this guaranty that all BLOB data inserted in the table to be compressed?
    Thank you for your help
    Taoufik

    The BLOB compression is a new feature of Oracle 11g.
    So why should it be possible to use this feature in a Oracle version which is about 5 years older? How should the DB be able to read compressed lob data if the lob compression feature is not implemented...

  • Anyone have on opinion on the usage of SECUREFILE LOBs for spatial data?

    Hi folks,
    From everything I read the usage of SECUREFILE LOBs over BASICFILE LOBs appears to be a nobrainer.
    http://www.oracle.com/technetwork/database/options/compression/overview/securefiles-131281.pdf
    Yet the default table creation remains set to create BASICFILE LOBs so my 11g spatial data is still sitting in BASICFILE LOBs (also I need to push it to and from 10g for the moment).
    Should we all be changing over to SECUREFILE LOB storage as a matter of course on 11g? Is there a big payoff for VARRAY storage beyond compression? A little payoff?
    Oracle says SECUREFILE LOBs "dramatically improve performance". Hey, I want some of that!
    http://www.oracle.com/newsletters/sap/products/database/oradb11g-features.html
    But yet there just has not been much traffic on the topic.
    Pro Oracle Spatial mentions weakly on page 250 that
    "Secure File LOBs are expected to be faster than BASIC LOBs".
    but seems to only be referring to the spatial index.
    Here Godfrind says to use them with points clouds but only trumpets the compression aspect.
    http://www.ncg.knaw.nl/Studiedagen/09PointClouds/presentations/PointCloud_14_AlbertGodfrind.pdf
    For those of us using ArcSDE, its pretty easy to add the SECUREFILE LOB keyword to the dbtune keyword that governs creation of the spatial column. But not a peep on the topic from the ESRI folks that I can find.
    Anyone have any experiences or opinions to share? I am hoping to finally leave 10g behind one day this year and putting together a migration plan. Should converting the LOBs be a first day bullet item or just something to consider sometime down the road?
    Thanks!
    Paul

    I have been writing on the benefits of rounding ordinates on storage and performance.
    That series has not yet been finished and fully published but my main finding (backed by stats/graphs) is that rounding ordinates does have an effect on performance as average coordinates per feature increases.
    As part of the work I also looked at SECUREFILE LOB storage in the hope that some sort of zlib/rle etc compression of the sdo_ordinates would substantially reduce storage size and thus increase performance.
    The reason I looked at this is because a customer has migrated from SDEBINARY to SDO_GEOMETRY (though most others are going from SDEBINARY to SDE.ST_GEOMETRY) and saw a substantial increase in storage costs. They also have taken a performance hit in some aspects of their operations but that may be due to ArcSDE programmer SDO_GEOMETRY access issues rather than any real issues with Oracle performance (though the perception is very much that Oracle is to blame). SDEBINARY was always a compressed storage format and with SDE.ST_GEOMETRY being WKB (of some description) also has a low storage footprint.
    The result of my work on LOB storage, though not published, is that I agree with Paul: there is little benefit for a lot of pain.
    However, I do still the whole question of storage size as being an area of benefit if reductions could be made as this is a necessary precursor for spatial being able to reap benefits if column-oriented storage ever migrates down to standard Oracle database.
    regards
    Simon

  • Storage for blob

    We have blobs that are small in size. If it is only 200 bytes will the storage be still one data block for that assuing we are storing all blocks with rest of the row. Please confirm. We are using 10g.

    Not 100% understand what you asked, however check Oracle Document for how LOB is stored.
    http://download-west.oracle.com/docs/cd/B19306_01/appdev.102/b14249/adlob_tables.htm#i1006363
    It can be outline or inline, by default LOB is stored out of line in seperate location from rest data of the Row.
    Or stored in line if you explicitly specify ENABLE STORAGE IN ROW for the LOB storage clause when you create the table

Maybe you are looking for