Sponsored Content
Top Forums UNIX for Advanced & Expert Users raid 0+1 +oracle getting slow data reads. Post 35969 by Optimus_P on Friday 16th of May 2003 08:34:51 AM
Old 05-16-2003
The machine has been setup and running for about two weeks.

Unfortunetly here at my work. they dont understand the importance of testing or a well planned upgrade stratigy.

These are my 4 steps to do over the weekend w/ this machine as far as fs tuening goes.

Code:
1) add the following entries to the /etc/system file (to enable larger cluster sizes then 128k)
*
* Allow larger SCSI I/O transfers, parameter is bytes (i can prolly scale this down to 512)
*
set maxphys = 1048576

*
* Allow larger DiskSuite I/O transfers, parameter is bytes (i can prolly scale this down to 512)
*
set md_maxphys = 1048576

2) set new cluster size to 512 for the FS (/dev/md/dsk/d100)
#RAID level 0, striping - Cluster size = number of stripe members x interlace size
tunefs -a 64 /dev/md/dsk/d100

3) set the ufs write throttle
*
* ufs_LW = 1/128th of memory
* ufs_HW = 1/64th of memory
*
set ufs_LW=16777216
set ufs_HW=33554432

4) disable fs caching edit /etc/vfstab and add option
nologging,forcedirectio

Here is the init.ora file
Code:
db_name = DEFAULT

db_file_multiblock_read_count = 8                                     # SMALL

db_block_buffers = 60                                                 # SMALL

shared_pool_size = 3500000                                            # SMALL

log_checkpoint_interval = 10000

processes = 50                                                        # SMALL

dml_locks = 100                                                       # SMALL

log_buffer = 8192                                                     # SMALL

sequence_cache_entries = 10                                           # SMALL

sequence_cache_hash_buckets = 10                                      # SMALL

max_dump_file_size = 10240      # limit trace file size to 5 Meg each

# Global Naming -- enforce that a dblink has same name as the db it connects to
global_names = TRUE

# Edit and uncomment the following line to provide the suffix that will be
# appended to the db_name parameter (separated with a dot) and stored as the
# global database name when a database is created.  If your site uses
# Internet Domain names for e-mail, then the part of your e-mail address after
# the '@' is a good candidate for this parameter value.

# db_domain = us.acme.com       # global database name is db_name.db_domain

# FOR DEVELOPMENT ONLY, DEFAULT TO SINGLE-PROCESS
# single_process = TRUE

# FOR DEVELOPMENT ONLY, ALWAYS TRY TO USE SYSTEM BACKING STORE
# vms_sga_use_gblpagfil = TRUE

# FOR BETA RELEASE ONLY.  Enable debugging modes.  Note that these can
# adversely affect performance.  On some non-VMS ports the db_block_cache_*
# debugging modes have a severe effect on performance.

_db_block_cache_protect = true                       # memory protect buffers
event = "10210 trace name context forever, level 2" # data block checking
event = "10211 trace name context forever, level 2" # index block checking
event = "10235 trace name context forever, level 1" # memory heap checking
event = "10049 trace name context forever, level 2" # memory protect cursors

# define two control files by default
control_files = (ora_control1, ora_control2)

 

7 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

Howto capture data from rs232port andpull data into oracle database-9i automatically

Hi, i willbe very much grateful to u if u help me out.. if i simply connect pbx machine to printer by serial port RS232 then we find this view: But i want to capture this data into database automatically when the pbx is running.The table in database will contain similar to this view inthe... (1 Reply)
Discussion started by: boss
1 Replies

2. Shell Programming and Scripting

Get data from 3 differrnt oracle DB & then compare data

Hi All, I have a requirement to write a shell script for the following... we have 3 different database.... lets say A, B, C From these 3 DBs, i need to get data.. all have 3 different table...a,b, c A.a => Emp_code, count(*) B.b => emp_code, count(*) C.c => emp_code, copunt(*) Once get this... (4 Replies)
Discussion started by: Amit.Sagpariya
4 Replies

3. Red Hat

Recover RAID data

Hello, Given a scenario, I have 2 HDD which were used on the server with software RAID. Now, the original server crashed and I have attached these 2 HDD to the new server. Any possible chances to recover the data from any of this HDD ? I want to mount /dev/sdb3 on some folder.. Output of... (3 Replies)
Discussion started by: chinmay
3 Replies

4. AIX

ORACLE Database running slow on AIX ( nmon / topas )

Hello, How can I know if ORACLE Database is running slow due to Memory or due to processing power ? I have only Oracle Database running on a P4 with 4GB RAM. Could anyone suggest any tools which can help me determine exactly if it is memory issue or processor issue. (43 Replies)
Discussion started by: filosophizer
43 Replies

5. AIX

SCSI PCI - X RAID Controller card RAID 5 AIX Disks disappeared

Hello, I have a scsi pci x raid controller card on which I had created a disk array of 3 disks when I type lspv ; I used to see 3 physical disks ( two local disks and one raid 5 disk ) suddenly the raid 5 disk array disappeared ; so the hardware engineer thought the problem was with SCSI... (0 Replies)
Discussion started by: filosophizer
0 Replies

6. UNIX for Dummies Questions & Answers

Difference between buffered disk reads and cached reads?

I was analyzing the Disk read using hdparm utility. This is what i got as a result. # hdparm -t /dev/sda /dev/sda: Timing buffered disk reads: 108 MB in 3.04 seconds = 35.51 MB/sec # hdparm -T /dev/sda /dev/sda: Timing cached reads: 3496 MB in 1.99 seconds = 1756.56 MB/sec... (1 Reply)
Discussion started by: pinga123
1 Replies

7. Red Hat

RAID Configuration for IBM Serveraid-7k SCSI RAID Controller

Hello, I want to delete a RAID configuration an old server has. Since i haven't the chance to work with the specific raid controller in the past can you please help me how to perform the configuraiton? I downloaded IBM ServeRAID Support CD but i wasn't able to configure the video card so i... (0 Replies)
Discussion started by: @dagio
0 Replies
post(3FORM)															       post(3FORM)

NAME
post - write or erase forms from associated subwindows SYNOPSIS
#include <form.h> int post_form(FORM *form); int unpost_form(FORM *form); DESCRIPTION
The function post_form displays a form to its associated subwindow. To trigger physical display of the subwindow, use refresh or some equivalent curses routine (the implicit doupdate triggered by an curses input request will do). The function unpost_form erases form from its associated subwindow. RETURN VALUE
These routines return one of the following: E_OK The routine succeeded. E_BAD_ARGUMENT Routine detected an incorrect or out-of-range argument. E_BAD_STATE Routine was called from an initialization or termination function. E_NOT_POSTED The form has not been posted. E_NOT_CONNECTED No items are connected to the form. E_NO_ROOM Form is too large for its window. E_POSTED The form has already been posted. E_SYSTEM_ERROR System error occurred (see errno). SEE ALSO
ncurses(3NCURSES), form(3FORM). NOTES
The header file <form.h> automatically includes the header file <curses.h>. PORTABILITY
These routines emulate the System V forms library. They were not supported on Version 7 or BSD versions. AUTHORS
Juergen Pfeifer. Manual pages and adaptation for new curses by Eric S. Raymond. post(3FORM)
All times are GMT -4. The time now is 09:33 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy