- Community Home
- >
- Servers and Operating Systems
- >
- Operating Systems
- >
- Operating System - HP-UX
- >
- Re: Oracle 9i performance
Categories
Company
Local Language
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Discussions
Discussions
Forums
Forums
Discussions
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
- BladeSystem Infrastructure and Application Solutions
- Appliance Servers
- Alpha Servers
- BackOffice Products
- Internet Products
- HPE 9000 and HPE e3000 Servers
- Networking
- Netservers
- Secure OS Software for Linux
- Server Management (Insight Manager 7)
- Windows Server 2003
- Operating System - Tru64 Unix
- ProLiant Deployment and Provisioning
- Linux-Based Community / Regional
- Microsoft System Center Integration
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Community
Resources
Forums
Blogs
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО04-05-2006 09:09 AM
тАО04-05-2006 09:09 AM
Re: Oracle 9i performance
In my experience, most Oracle performance problems aren't caused by kernel settings. For the most part, kernel tweaking is a matter of simply giving Oracle the resources it wants. If Oracle doesn't have those resources, its not that it runs slowly, it simply doesn't run at all. It fails with a failure to allocate some IPC resource or fork a process or whatever. Although I have seen I/O get hung for extreme IO loads if the scsi_queue_depth setting was too low on fibre adapters.
Likewise, few performance issues are caused by Oracle parameter settings, for the same reasons. For the things that can be checked, make sure that db_cache_size and shared_pool are high enough. Also make sure that the database stats are analyzed about once a week.
The best place to start looking is within the database; and the best tool to start looking with is statspack. It will identify the top wait events within the database, as well as the SQL statement causing the highest disk I/O. I have seen a lot of performance problems caused by SQL that was not properly optimized for Oracle.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО04-05-2006 10:56 PM
тАО04-05-2006 10:56 PM
Re: Oracle 9i performance
also check in your Database control (Enterprise Manager), you have many advisors to help as well as ADDM.
hope this helps too!
kind regards
yogeeraj
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО04-06-2006 05:13 AM
тАО04-06-2006 05:13 AM
Re: Oracle 9i performance
I don't have access to the database and hence this question. I am contributing by reading and imagination.
Thanks,
Shiv
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО04-06-2006 06:32 AM
тАО04-06-2006 06:32 AM
Re: Oracle 9i performance
Why do you think or claim that you have a Disk I/O performance issue? I say - first establish that you indeed have an I/O problem. And this is 2 pronged:
(1) DBA approach -- ask your DBA to ascertain if there really is an obvious I/O problem. Ask what volumes/filesystems/files suffer from "slow" performance.
(2) On your end - SysAdmin, check the volume's/filesystem's in question -- check the component disks via "sar -d" if you indeed have I/O issues -- which should manifest via qdepths > 1.0
If (1) or (2) confirms you indeed have an I/O issue, then look at how your XP1024 LUNs are used to build your Oracle storage. It does not matter whether you use RAW or cooked storage -- best pracice for the XP array is to always stripe accross 4 or 8 LUNs with each LUN coming from different ACP and array group and prefereably on each own Fibre Channel Path...
Follow the above recipe and you will alwyas have proof that your I/O configuration is not at fault..
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО04-06-2006 07:48 PM
тАО04-06-2006 07:48 PM
Re: Oracle 9i performance
more information on statspack is available at:
http://www.oracle.com/technology/deploy/performance/pdf/statspack.pdf
if you have further queries, please do let us know
kind regards
yogeeraj
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО04-07-2006 01:14 AM
тАО04-07-2006 01:14 AM
Re: Oracle 9i performance
Some basic stats you should monitor:
select (bbc.total_waits*100/(cg.value+dbg.value)) "Buff busy ratio ind. <= 0,007"
from v$system_event bbc,
v$sysstat cg, v$sysstat dbg
where bbc.event='buffer busy waits'
and cg.name ='consistent gets'
and dbg.name='db block gets';
select (1-(sum(getmisses)/sum(gets))) * 100 "Hit Ratio > 95%"
from v$rowcache;
select round((congets.value+dbgets.value-physreads.value)*100/
(congets.value+dbgets.value),4) || '%' "Buffer Cache Hit Ratio >= 95%"
from v$sysstat congets, v$sysstat dbgets, v$sysstat physreads
where congets.name='consistent gets'
and dbgets.name='db block gets'
and physreads.name='physical reads';
select round(100 * sum(reloads) / sum(pins), 2) "Reloads Pct < 1%"
from v$librarycache;
Best Regards,
Eric Antunes
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО08-12-2006 06:45 AM
тАО08-12-2006 06:45 AM
Re: Oracle 9i performance
I agree with one of the posts above about Solid state disk to an extent. It is a great solution for I/O issues, however you do not need to put your entire dbase on the SSD. So it can be affordable.
I own a company that sells SSD SANS, and the added benefit of putting your most accessed data on them is that they are no longer accessed from the array...making the array work more efficiently as well. We have found that usually 2-5% of your data can cause 80%-90% of your IO problems. If you would like to learn more, please send me an e-mail mike@bdtstorage.com.
MLK
- « Previous
-
- 1
- 2
- Next »