I will be installing 11.20.3 Database in Aix 6.1 - I'am worry about the fs blocksize and redo blocksize
Hi experts.
I will be installing 11.20.3 Database in Aix 6.1 - I'am worry about the fs blocksize and redo blocksize.
We have a lot of db's with I/O problems and maybe the way the filesystems are created can be a issue?
Also I wonder if a redo of 4k can be more efficient than the default of 512?
We have a similar db in 11.2 and also Ibm Aix 6.1 and this is the wastage:
SQL> conn system
Enter password:
Connected.
SQL> SELECT BLOCKSIZE FROM V$LOG;
BLOCKSIZE
----------
512
512
512
SQL> SELECT name, value FROM v$sysstat WHERE name = 'redo wastage';
NAME VALUE
--------------------------
redo wastage 1386749364
I will be installing 11.20.3 Database in Aix 6.1 - I'am worry about the fs blocksize and redo blocksize.
We have a lot of db's with I/O problems and maybe the way the filesystems are created can be a issue?
Also I wonder if a redo of 4k can be more efficient than the default of 512?
We have a similar db in 11.2 and also Ibm Aix 6.1 and this is the wastage:
SQL> conn system
Enter password:
Connected.
SQL> SELECT BLOCKSIZE FROM V$LOG;
BLOCKSIZE
----------
512
512
512
SQL> SELECT name, value FROM v$sysstat WHERE name = 'redo wastage';
NAME VALUE
--------------------------
redo wastage 1386749364
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
OK, "cio" is not in the list of the FS options. I've asked, because combination of "cio" and Oracle file autoextending could cause massive performance problems on AIX.
Anyway, looks like this is not Your case.
Consider about increasing the block size at least to 4K.
Anyway, looks like this is not Your case.
Consider about increasing the block size at least to 4K.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
See my previous comment/post
ASKER
node mounted mounted over vfs date options
-------- --------------- --------------- ------ ------------ ---------------
/dev/hd4 / jfs2 Apr 12 12:09 rw,log=/dev/hd8
/dev/hd2 /usr jfs2 Apr 12 12:09 rw,log=/dev/hd8
/dev/hd9var /var jfs2 Apr 12 12:09 rw,log=/dev/hd8
/dev/hd3 /tmp jfs2 Apr 12 12:09 rw,log=/dev/hd8
/dev/hd1 /home jfs2 Apr 12 12:09 rw,log=/dev/hd8
/dev/hd11admin /admin jfs2 Apr 12 12:09 rw,log=/dev/hd8
/proc /proc procfs Apr 12 12:09 rw
/dev/hd10opt /opt jfs2 Apr 12 12:09 rw,log=/dev/hd8
/dev/livedump /var/adm/ras/livedump jfs2 Apr 12 12:09 rw,log=/dev/hd8
/dev/martdev_orac_lv /oracle jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/martdev_data_lv /martdev/DATA jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/martdev_indx_lv /martdev/IDX jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/martdev_red1_lv /martdev/REDO1 jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/martdev_red2_lv /martdev/REDO2 jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/martdev_red3_lv /martdev/REDO3 jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/martdev_syst_lv /martdev/SYSTEM jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/martdev_temp_lv /martdev/TEMP jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/martdev_undo_lv /martdev/UNDO jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/martdev_dump_lv /dump_ora/martdev jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/martdev_arch_lv /martdev/ARCHIVES jfs2 Apr 16 09:39 rw,log=/dev/loglv00
/dev/backup_lv /backup jfs2 Apr 18 11:18 rw,log=/dev/loglv00
datamartdev-oracle:/home/o