| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
 |
 | |||
| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
|
| |||
Home -> Community -> Mailing Lists -> Oracle-L -> RE: Bitmap Header in a Uniform LMT?
Thanks a ton Dennis. I will certainly go thru this. It looks more detailed.
I also found a much similar analysis at
http://www.performance-insight.com/ora3_back.html
Raj
DENNIS WILLIAMS
<DWILLIAMS_at_LIFE To: Multiple recipients of list ORACLE-L <ORACLE-L_at_fatcity.com>
TOUCH.COM> cc:
Sent by: Subject: RE: Bitmap Header in a Uniform LMT?
root_at_fatcity.co
m
August 09, 2002
04:53 PM
Please respond
to ORACLE-L
Raj - My apologies for being late with a reply. Your latest message
prompted
me to recall that Jeremiah Wilton investigated this (or a similar aspect)
back in April. I went into Google and typed "LMT bitmap headers" (without
the quotes), and it retrieved the discussion. I have inserted it below for
your review in case you were not aware of that discussion.
Dennis Williams
DBA
Lifetouch, Inc.
dwilliams_at_lifetouch.com
Received: from newsfeed.cts.com (newsfeed.cts.com [209.68.248.164])
by naude.co.za (8.11.2/8.11.2) with SMTP id g38Lpc327439
Received: from fatcity.UUCP (uucp_at_localhost)
by newsfeed.cts.com (8.9.3/8.9.3) with UUCP id OAA08313;
Mon, 8 Apr 2002 14:58:27 -0700 (PDT)
Received: by fatcity.com (26-Feb-2001/v1.0g-b71/bab) via UUCP id 0043EF61;
Mon, 08 Apr 2002 13:03:20 -0800
Following on from my previous note:
Jeremiah,
>From your bitmap control,
You have FF occurring 3 times followed by 3F which is
255, 255, 255, 63 which is
11111111 11111111 11111111 00111111
So, least signficant bit first,
11111111 11111111 11111111 11111100 which is
used, used, ... (30 times) , free, free
This corresponds with the first: 30 (the bit before the first free bit)
Paul
-----Original Message-----
Sent: Monday, April 08, 2002 3:23 PM
To: 'ORACLE-L_at_fatcity.com'
>From the 'Data Management and Storage Internal" notes,
Bitmapped Tablespace File Structure
A new bitmapped tablespace file has the following structure:
File Header 1 block Bitmapped File Space Header 1 block Head portion of of Bitmap Blocks N blocks Useful file blocks U units (A unit isa number
Tail portion of Bitmap Blocks M blocks
If a Unit = B blocks, then the total file size = 1 + 1 + N + U*B + M
Bitmapped File Space Header
...... (lots to type, I can if you really need it)
Bitmap blocks have 2 parts :
Bitmap control structure
Vector Dump
The fields in the bitmap control structure are:
RelFNo: Relative file number to which the bitmap belongs BeginBlock: Which block number does the first bit represent Flag: Zero for permanent files, one for temp files First: Where to start looking for the free space (bit before firstfree
Free: Number of free slots (bits) in the bitmap (notthe file)
To read the bitmap, take each two-byte pair, least significant bit first.
If there are not eight bits, pad to eight bits with zeroes. Hence 0x0F =
15
= 00001111. When written least significant bit first, the bitmap looks
like
this
11110000 --> used, used, used, used, free, free, free, free
Scanning for the first free extent will start at the 4th bit.
HTH Paul
-----Original Message-----
Sent: Monday, April 08, 2002 3:44 PM
To: Multiple recipients of list ORACLE-L
Out of curiosity I decided I wanted to look at what composed the extent map in locally-managed tablespaces.
I dumped the first 5 blocks of the tablespace's first datafile with 'alter system dump datafile ...' The results surprised me, as they appeared to consist of almost no data. The LMT in question contains a variety of segments and extents. How is the LMT bitmap organized?
Start dump data blocks tsn: 1 file#: 2 minblk 1 maxblk 1 Block 1 (file header) not dumped: use dump file header command
Start dump data blocks tsn: 1 file#: 2 minblk 2 maxblk 2
frmt: 0x02 chkval: 0x0000 type: 0x1d=KTFB Bitmapped File Space Header
File Space Header Block:
Header Control:
RelFno: 2, Unit: 8192, Size: 524352, Flag: 1
Initial Area: 3, Tail: 524292, First: 30, Free: 34
Start dump data blocks tsn: 1 file#: 2 minblk 3 maxblk 3
frmt: 0x02 chkval: 0x0000 type: 0x1e=KTFB Bitmapped File Space Bitmap
File Space Bitmap Block:
BitMap Control:
RelFno: 2, BeginBlock: 5, Flag: 0, First: 30, Free: 128994
FFFFFF3F00000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000... all zeros
Start dump data blocks tsn: 1 file#: 2 minblk 4 maxblk 4
frmt: 0x02 chkval: 0x0000 type: 0x1e=KTFB Bitmapped File Space Bitmap
File Space Bitmap Block:
BitMap Control:
RelFno: 2, BeginBlock: 1056964613, Flag: 0, First: 0, Free: 129024
0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000... all zeros
FWIW: SQL> select count (*) from dba_extents where file_id = 2;
COUNT(*)
30
SQL> select extent_management from dba_data_files df, dba_tablespaces ts where df.tablespace_name = ts.tablespace_name and file_id = 2;
EXTENT_MAN
-- Jeremiah Wilton http://www.speakeasy.net/~jwilton -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Jeremiah Wilton INET: jwilton_at_speakeasy.net Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists -------------------------------------------------------------------- -----Original Message----- Sent: Friday, August 09, 2002 3:33 PM To: Multiple recipients of list ORACLE-L Getting no responses on this, I created a test 9.0 database on my NT 4.0 workstation, with a block size of 4K. My observations are : While creating a LMT with uniform extents of size 16K, the minimum datafile size has to be 25K. With uniform extents of size 32K, the datafile size has to be more than 41K, and with uniform extents of size 64K, the minimum datafile size is 73K. Thus, the size of the datafile should be atleast 2 blocks more than the uniform extent size. SQL> create tablespace mult 2 datafile 'd:\mult7.dbf' size 10M, 3 'd:\mult2.dbf' size 41K, 4 'd:\mult3.dbf' size 73K, 5 'd:\mult4.dbf' size 128K, 6 'd:\mult5.dbf' size 1024K, 7 'd:\mult6.dbf' size 2048K 8 extent management local uniform size 32K 9 / Tablespace created. SQL> select * from dba_free_space where tablespace_name='MULT'; TABLESPACE_NAME FILE_ID BLOCK_ID BYTES -------------------- ---------- ---------- ---------- ---------- ------------ MULT 9 17 10420224 2544 MULT 10 4 32768 8 MULT 11 4 65536 16 MULT 12 17 65536 16 MULT 13 17 983040 240 MULT 14 17 2031616 496 6 rows selected. And right now, I have the answer to all my questions. Also, I strongly recommend visiting http://www.performance-insight.com/ora3_back.html and look at the topics on locally managed tablespaces. I was impressed. That was what helped me get to the right answers. Regards Raj Rajesh Rao/CHASE_at_CHA To: Multiple recipients of list ORACLE-L <ORACLE-L_at_fatcity.com> SE cc: Sent by: Subject: Bitmap Header in a Uniform LMT? root_at_fatcity. com August 09, 2002 02:28 PM Please respond to ORACLE-L I asked this question the other day, but got no replies. I make one more attempt and rephrase the question. A curious mind wants to know " Assuming a block size of 2K, what would be the size of the bitmap header if I create a uniform LMT with datafiles of size 32K, 64K, 100K, 1M and 64M?" Metalink has not given me any definite answer as yet, and the notes there have confused moi. Regards Raj -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: INET: Rajesh.Rao_at_jpmchase.com Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists -------------------------------------------------------------------- To REMOVE yourself from this mailing list, send an E-Mail message to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: DENNIS WILLIAMS INET: DWILLIAMS_at_LIFETOUCH.COM Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists -------------------------------------------------------------------- To REMOVE yourself from this mailing list, send an E-Mail message to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: INET: Rajesh.Rao_at_jpmchase.com Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists -------------------------------------------------------------------- To REMOVE yourself from this mailing list, send an E-Mail message to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).Received on Fri Aug 09 2002 - 16:38:40 CDT
 |
 |