HP Unix memory tuning

Hi.
i want to increase PXA buffer size in ECC 5.0 -Netweaver 2004
OS -HP Unix
Ram:16GB,Number of swaps per day very high.
Existing size:220MB .Will it result in any OS related problems
Is PXA Parameter dependant on any oher sap paramters?
Regards
Arun

Dear Arun,
Some recomendations for this buffer area are given in note 103747.
I recommend to you after increasing the size, run sappfpar command on operating system level
in order to detect any memory errors or warnings.
If number of swaps are high, it could be a good idea to increase it, or to move to a faster drives.
Hope it helps.
Edited by: alejovw on Mar 1, 2011 9:08 PM

Similar Messages

Memory tuning

Any memory tuning to improve window 7 enterprise performance? Pls advice

Hey Sam,
There are some things that can improve performance, because I wrote things that moderators do not have the specs of the computer and the software installed.
Cooling is important to have a good enough computer, and check the temperature of the processor (Fans often recommended to replace it incredibly improves performance)
Check out time access I / O hard disk (using Disk HIRENS possible) and check the disk responds correctly (50-100) and no BAD SECTOR.
Consider replacing the hard disk type SSD (much faster).
Some of the installations intelligently (often prefer a clean OS install on a single drive, and a different drive programs to improve performance).
The file FILE PAGE Put on your operating system is different.
The PAGE file FILE Put a different operating system drive and limit it according to Microsoft's recommendations (usually between 1.5-3 size of RAM).
RAM memory is recommended at least 8 GB, and check that the operating system supports (64 bit)
Remove the effects that operating system kit (fade effect & aero theme)
Please Mark This As Answer if it helps to solve the issue
Tzuri Ben Ezra | My Certifications:
CompTIA A+ ,Microsoft MCP, MCTS, MCSA, MCITP
|
FaceBook: Tzuri FaceBook | vCard:
Tzuri vCard |
Microsoft ID:
Microsoft Transcript
|

Memory Tuning - SAPPFPAR check

I was wondering on how to optimize the use of "SAPPFPAR check" in terms of Memory Tuning. Upon typing the command "sappfpar check pf=<profile>", the shared memory disposition overview is being shown. Are the values shown here present the current values defined in the system?
Furthermore, i was wondering what values are shown below? Do all the values shown below display the currently allocated values in the system or are the values shown below display the required value at the very least? If so, how can i change the value? How do i know what parameter these values are directly corresponding to so that i can check and change the said values if necessary? Especially the values shown in the "worst case" field, how can i check whether the currently allocated values satisfy what are indicated in the SAPPFPAR check?
In addition, i was wondering what does the error imply? Does it straightforwardly imply that there are incorrect values existing in the system, or does it imply that requirements are not met, or etc. I would like to know more about this from the SAP practitioners themselves since experience provide the best knowledge that can be taught since it occurred first hand.
Kindly enlighten me on this topic. Thanks.
Edited by: Jett Michael Contreras on Apr 22, 2008 11:36 AM

Hi
The SAPPFPAR checks the current values that is mentioned in the instance profile . Unless if u have made changes to the instance profile. If u do make changes to the profile then the SAPPFPAR will show the result for the changed profile.
The output given by the SAPPFPAR will show you the current value and the required value (ie Proposed SAP values) to be changed. If there is any error in the profile parameters it indicates that in the output. Eventually it calculates all the SAP buffers , Physical memory and the SWAP space. It gives you the total required memory by adding the worst case scenario (heap) alongwith the buffer parameters.
When u change the buffer parameters for the system, it will calculate and adjust according to the SAP proposed value. Since the shared memory values depends on all the buffer parameters like program, screen, table buffers..etc.
The SAPPFPAR is a very helpful tool for calculating the physical memory space required based on the SAP buffer and for checking error in the instance parameter.
Thanks & Regards,
Maheswaran

OEM in Linux and Unix Memory difference on Deployed OC4J Container

I have 2 Oracle 9iAS environments 1 each for production and test.
My Prod servers are on AIX (Unix)
My Test server is on Linux
Under Enterprise Manager If I create an OC4J container exactly the same in both and deploy an application exactly the same in both could someone tell me why the memory usage display is widely different - on linux its much more used than on unix?
Perplexing.

The code gets pretty messy as I have both a SendMail and MailMessage class that I use. Basically for not authenticated messages (I need authenticated), the example on metalink Doc ID: Note:234490.1 with subject "Mail Service Setup" is what I used to get this going.
I updated the local application.xml file located in C:\jdev1012\jdev\system10.1.2.2.0.1929\oc4j-config\application.xml
<mail-session location="mail/mailsession" smtp-host="my.mail.server">
<property name="mail.transport.protocol" value="smtp"/>
</mail-session>
On a datapage, I have a button with event_Email. In the action class I have an onEmail() method that calls the code to create a message and send it with the email transport.
In my libraries in JDeveloper, I created one called JavaMail which has activation.jar and mail.jar. This is added to the project.
Everything works fine when running locally. The problem is in deployment. The OC4J container that I deploy to has a java2.policy file. At first the classpath, and the location in the policy file for mail.jar and activate.jar were not the same. I updated the policy file to use the files in the classpath. Since mail.jar and activate.jar on the OAS server are in the classpath, I didn't include these files in the deployment.
When trying to run, I don't get any messages, just a 500 Error.
If I try to implement a policy file on the local OC4J in JDeveloper, I get lots of problems, and I gave up on that route.
Hope that adds more info. Thanks, Ken

HP UNIX memory problems

Hi While upgrading to 6.2 on HP Unix we are getting the follwing error during the outline build.Unable to Allocate Aligned Memory for [pMemByte] in [adDatGetFreeFrame].Any ideas or solutions?Thanks for your help.Srini

HiI worked with a HP N4000 with 6 CPUs and 6 GB of RAM used OS HP-UX 11.x, the problem with the memory is related with the kernel configuration, you need to set some parameters to allow the kernel assign more address of memory for Essbase or another applicationOmar Aecio G. R.Business Intelligence ConsultantOmniSys S.A. de C.V.e-mail: [email protected]

Need help with performance & memory tuning in a data warehousing environment

Dear All,
Good Day.
We had successfully migrated from a 4 node half-rack V2 Exadata to a 2 node quarter rack X4-2 Exadata. However, we are facing some issues with performance only for few loads while others have in fact shown good improvement.
1. The total memory on the OS is 250GB for each node (two compute nodes for a quarter rack).
2. Would be grateful if someone could help me with the best equation to calculate the SGA and PGA ( and also in allocation of shared_pool, large_pool etc) or whether Automatic Memory Management is advisable?
3. We had run exachk report which suggested us to configure huge pages.
4. When we tried to increase the SGA to more than 30GB the system doesn't allow us to do so. We had however set the PGA to 85GB.
5. Also, we had observed that some of the queries involving joins and indexes are taking longer time.
Any advise would be greatly appreciated.
Warm Regards,
Vikram.

Hi Vikram,
There is no formula about SGA and PGA, but the best practices for OLTP environments is for a give ammount of memory (which cannot be up to 80% of total RAM from server) you should make 80% to SGA and 20% to PGA. For Data Warehouse envs, the values are like 60% SGA and 40% PGA or it can be up to 50%-50%. Also, some docs disencourage you to keep the database in Automatic Memory Management when you are using a big SGA (> 10G).
As you are using a RAC environment, you should configure Huge Pages. And if the systems are not allowing you to increase memory, just take a look at the semaphore parameters, probably they are set lower values. And for the poor performance queries, we need to see explain plans, table structure and you would also analyze if smart scan is playing the game.
Regards.

Windows XP 32-bit Virtual Memory Tuning.

For our PXI racks we "tune" the virtual memory under Windows XP 32-bit (standard NI installation), as a couple of our applications process vast volumes of data , as follows:
Open boot.ini
Find the line beginning with 'multi'
Add the following to the end of the line '/3GB /userva=2560' (without the quotes).
Initially we were supplied with PXIe-8133 driven PXI racks and the above change works perfectly.
However later PXI racks were supplied with PXIe-8115 driven PXI racks. If the above line is applied to these PXI racks then the result is the graphics card drops to 640x480 4-bit colours and it cannot be changed until the line is removed.
I have tried various values for /userva including 2176, 2432, and 2304 but all results in the same result.
It seems that whilst the PXIe-8133 came with a standard BIOS and hence was configurable to an extent, the PXIe-8115 comes with a NI BIOS and provides very little low level control in comparison.
Has anyone attempted the above on a PXIe-8115 driven rack? Or can suggest a solution?
Thanks.
Christopher Povey
Senior Test Systems Engineer for BAE Systems.

I have attached the Windows XP System Resource Maps for both our PXIe-8115 and PXIe-8133 racks. Both racks contain the same hardware in the same slots (apart from the controller card).
Christopher Povey
Senior Test Systems Engineer for BAE Systems.
Attachments:
PXIe-8115 Memory Resources.txt ‏11 KB
PXIe-8133 Memory Resources.txt ‏8 KB

Memory tuning imap store

Hello,
i m trying to tune my web application, a web mail, once a user logs in I keep an IMAPStore instance inside the httpsession.
I noticed it takes 6 mb of memory right away, just after connecting to the imap store, once I open a few folder and fetch a few small messages it goes up to 8 or 9 mb.
to get messages i follow this steps:
- openFolder
- fetchmessages
- populate my beans
- invalidateheaders for each imapmessage to release cached data
- call folder.clearCache() (i added this method inside imapfolder to clear messageCache vector)
- close folder
but memory usage is still the same.
Is there anything to reduce memory usage? i should have no more than 1MB per session.
Thank you.

Hi,
I am using MessAdmin to measure HttpSession data.
6mb is the size of IMAPStore, and this size appear even if i don't open a folder, just after getting the store:
session = Session.getInstance(props, auth);
session.setDebug(debug);
store = (IMAPStore) session.getStore(imap);
store.connect();session and store object measure 6mb.
If you have a better tool please let me know.
thank you

HADB Memory tuning?

Hi all,
We have installed ONLY HADB component from SJSAS 9.1. As we are using Java 1.5 version, the default jvm heap size is 192m which is way less than our requirement. How can I tune the JVM used by HADB? I saw that the service clu_nsup_srv running under JVM process.
Thanks,
Paras

Hello Parasshah,
You might want to post HADB related questions to the App Server forum. Here is the link to that forum:
[ Sun Java System Application Server - Performance and Tuning|http://forums.sun.com/forum.jspa?forumID=756]

Performance Tuning 10g

Hi All,
I had given a task to tune oracle 10g database. I am really new in memory tuning although I had some SQL Tuning earlier. My server is in remote location and I can not login to Enterprise Manager GUI. I will be using SQL Developer or PL/SQL Developer for this. My application is web based application.
I have following queries with this respect:
- How should I start... Should I use tkprof or AWR.
- How to enable these tools.
- How to view its reports
- What should I check in these reports
- Will just increasing RAM improves performance or should we also increase Hard Disk?
- What is CPU Cost and I/O?
Please help.
Thanks & Regards.

dbdan wrote:
Hi All,
I had given a task to tune oracle 10g database. I am really new in memory tuning although I had some SQL Tuning earlier. My server is in remote location and I can not login to Enterprise Manager GUI. I will be using SQL Developer or PL/SQL Developer for this. My application is web based application.
I have following queries with this respect:
- How should I start... Should I use tkprof or AWR.
- How to enable these tools.
- How to view its reports
- What should I check in these reports
- Will just increasing RAM improves performance or should we also increase Hard Disk?
- What is CPU Cost and I/O?
Please help.
Thanks & Regards.Here is something you might try as a starting point:
Capture the output of the following (to a table, send to Excel, or spool to a file):
SELECT
STAT_NAME,
VALUE
FROM
V$OSSTAT
ORDER BY
STAT_NAME;
SELECT
STAT_NAME,
VALUE
FROM
V$SYS_TIME_MODEL
ORDER BY
STAT_NAME;
SELECT
EVENT,
TOTAL_WAITS,
TOTAL_TIMEOUTS,
TIME_WAITED
FROM
V$SYSTEM_EVENT
WHERE
WAIT_CLASS != 'Idle'
ORDER BY
EVENT;Wait a known amount of time (5 minutes or 10 minutes)
Execute the above SQL statements again.
Subtract the starting values from the ending values, and post the results for any items where the difference is greater than 0. The Performance Tuning Guide (especially the 11g version) will help you understand what each item means.
To repeat what Ed stated, do not randomly change parameters (even if someone claims that they have successfully made the parameter change 100s of times).
You could also try a Statspack report, but it might be better to start with something which produces less than 70 pages of output.
Charles Hooper
IT Manager/Oracle DBA
K&M Machine-Fabricating, Inc.

Can we ensure memory is released from bpel , to be picked by GC

Dear All,
We are in production support senario.
We are doing memory intensive operations using BPEL.
We are looking for ways to optimize db adapter peformance.
We are looking for ways to ensure that memory is released by bpel after operations.
Please suggest
Regards
Arc

hmmm - if you'd be a little more specific - we should be able to help you better. Is there anything un-normal that you see? a heap dump might reveal if there are options to towards mem tuning.
also what is the version of bpel that you are using.
--- A couple of thoughts regarding memory tuning.
a) keep variables scoped - and not global - that helps us to collect / recycle them faster - especially during a while loop
b) in some cases pass by reference should help (skipDeepCopy property on the partnerlink) - by default we do a copy by value
c) if you use stylesheets (xsl) there is a known xml parser leak - patch downloadable from metalink.
cheers

Tuning Oracle10g

Hi All,
I am learning Oracle Memory Tuning and gone through some Performance Tuning Articles. My questions are:
- Since Oracle10g has Automatic Memory Management, once we specified Max SGA and MAX PGA what we need to tune in memory?
- Where can I get tuning examples with case scenarios?
Thanks & Regards,
Danish

Hi Danish,
Since Oracle10g has Automatic Memory Management, once we specified Max SGA and MAX PGA what we need to tune in memory?My company supports hundreds of databases, and I've noticed that while this "autoimated" memory management is OK for small databases, it may not be right for shops with a large or busy system. This is because AMM is "reactive", and waits until it detects a problem before addressing it. I have an alert here about it, how AMM resise operations can cause performance issues, in some cases:
http://www.dba-oracle.com/t_amm_automatic_memory_resize.htm
Oracle has an insatiable appetite for RAM, but there are diminishing marginal returns as you approach full caching.
Your goal is to give him enough RAM to hit the "sweet spot" of optimal pool sizes.
Oracle has pool advisory utilities that can assist you in this, check them out . . . .
Also, I wrote this note especially for people like you, who use AMM and want to tune RAM:
http://www.oracle-training.cc/t_amm_oracle_instance_tuning.htm
Since you are on 10g, I've collected numerous "gotchas" in 10g that you may find useful:
http://www.dba-oracle.com/t_slow_performance_after_upgrade.htm
Where can I get tuning examples with case scenarios?There are literally thousands of examples on the web. I would search like this, for Oracle RAM (or SGA) tuning examples:
http://www.google.com/search?&q=oracle+ram%7Csga+tuning+example
In sum, the best way to learn how to optimize RAM is to dive-in and get your feet wet!
Hope this helps . . .
Donald K. Burleson
Oracle Press author
Author of "Oracle Tuning: The Definitive Reference"
http://www.rampant-books.com/book_2005_1_awr_proactive_tuning.htm

Reg Instance Tuning

Hi Pals,
Can anyone help me out in 'Instance tuning'.The situation is like this- Bulk of simulated users hit the database and I need to tune the DB Oracle8.1.7
some points for consideration are CPU utilization,Memory tuning and contention.
Wat couple of parameters atleast u guys out there suggest to be considered for tuning?
I would appreciate pointers-books/links/tools to achieve this and to get more info and guidance.
Thanks and cheers,
sreek

You can go for tuning db block buffers, shared pool size at instance level. You can also tune short area size. However make sure that sum of memory allocated to sga does not exceed 50% of RAM. You can also go for application level tuning to improve ur database performance

LEARNING ORACLE PERFROMANCE TUNING

Hi everyone,
Can any one guide me how to learn Performance tuning with examples becoz when i follow the oracle doc's.It is little bit diffcult for me to understand.
Regards,
Vikas Kohli
Edited by: Vikas Kohli on Aug 30, 2011 3:40 PM

Hi,
We will increase the database performance by tuning sql queries, instance memories, by reading awr report.
In sql tuining, have to tune sql quries based on how much response time is taking to get output and how much cpu is using. this we will check by enabling autotrace for query. oracle will choose the best execution plan of sql statement base on statistics that are available on the objects.
In memory tuning buffer cache, redo buffer, shared pool memories has to tune to reside the data in the buffer cache without waiting for free buffers.
In 10g and 11g oracle only automatically adjust the memories with the limit of sga_target parameter value. Even though we want we can tune memories manually.
AWR report mainly gives information about health of the database. how database is performing. that means how the database is responding to users. how sql statements are getting result. where that statements waiting to get result. is that waiting for buffer cache, or waiting to get data from disk, is that is reading lots of indexes etc. like that.
I hope you will get clear idea now.
thanks

Is 500mb of swap used normal when there is still free memory?

Kind of worried First mac so might be abit new to OS X. Thanks in advance!

Kev~* wrote:
Right now I still have a relatively small number of apps that are open (~8-9), which doesnt justify such high ram usage, much less 500mb of swap.
The number of applications actually doesn't matter one bit. It's the type of applications and what kind of RAM they need. I have read some Mac engineers say that some applications need to preallocate RAM because of what they do, although OS X provides for ways to do that which will allow that RAM to be released to other applications if needed.
I could open 12 consumer apps and not see much change in swap file usage. But I could reboot and open just Photoshop and have it make a panorama or HDR of several DSLR images, or fire up a single 3D, motion graphics, or scientific program, and RAM/swap file requirements could instantly outpace anything general office apps could ever need.
Since you're new to OS X, you might want to become familiar with the UNIX memory management model that OS X uses. In other words, RAM management in OS X may be unlike what you were used to before, but its UNIX-based model is normally very efficient, especially with the new Mavericks RAM compression.
Also, based on what I've seen (using graphics apps), only 500MB swap wouldn't even get my attention...

HP Unix memory tuning

Similar Messages

Maybe you are looking for