Buffer Parameter Increase???

In one of the servers we are facing high number of swaps in the following buffers.
Program buffer                   (PXA)
CUA buffer                       (CUA)
Screen buffer                    (PRES)
Export/Import buffer             (EIBUF)
To overcome this issue we have to increase the value in corresponding
Program buffer                   (PXA)
abap/buffersize
CUA buffer                       (CUA)
rsdb/cua/buffersize
Screen buffer                    (PRES)
zcsa/presentation_buffer_area
Export/Import buffer             (EIBUF)
rsdb/obj/buffersize
My Question is how do I analyze how much Value should I increase for each of the Parameter.
Useful answers will be rewarded!

Hello,
In st02, double click on the parameter buffer that faces many swap, for exemple "Screen".
The following tab show you
Size              Allocated        KB
                  Available        KB
                  Used             KB
                  Free             KB
Directory entries Available
                  Used
                  Free
This buffer is setup by 2 parameters (you can display them with 'current parameter button'):
zcsa/presentation_buffer_area
sap/bufdir_entries
One parameter is an amount of memory (in Bytes), the other represent a 'maximum number'.
To know if you need more memory place in the buffer or to increase the max number, you need to read the tab on the previous screen.
If the "Size" - "Free" is next to 0, you need to increase zcsa/presentation_buffer_area
If the "Directory entries"-"Free" is next to 0, you need to increase sap/bufdir_entries
There is no absolute rule to know in one row how much you need to increase every parameters for every buffer. It's more a question of 'fellings' and you often need to do many tests (change parameter, stopsap/startsap, re-change parameter, etc.).
For exemple
zcsa/presentation_buffer_area = 10000000
sap/bufdir_entries = 10000
If "Size"-"Free" is to 0 and "Directory entries"-"Free" is to 9000, this means you can have 9 times more directory entries but no more memory to hold them.
So, on this system 1000 buffer entries equals to 10Mb of memory.
So if "Objects swapped" is to 2000, just set zcsa/presentation_buffer_area = 10000000 +2*10000000
You don't have to modify sap/bufdir_entries because it can handle 9 more time entries.
If you want to handle in memory as much entries as the maximum directory entries, you could set zcsa/presentation_buffer_area = 10000000 + 9*10000000
Do the same way for every buffer with many swaps.
Hope this will help you.
Regards,

Similar Messages

Recordlength & buffer parameter in IMP

In IMPORT of oracle we specify 2 parameter
1. recordlength
2. buffer
1. How much maximum size we can spercify for those parameter to fasten the IMPORT process? Also on which thing those values depends?
2. Also how much time will it save if we specify those parameters? e.g. If any IMPORT takes 10 Hours to import and if we specify those 2 parameters to high values then how much will it take to import?
3. Is there any other parameter which will fasten the import process?
Thanks,

for first You have to understand what those parameters mean:
The BUFFER parameter applies ONLY to conventional path Export. It has no effect on a direct path Export. This BUFFER parameter specifies the size (in bytes) of the buffer used to fetch rows. It determines the maximum number of rows in an array, fetched by Export. For direct path Export, use the RECORDLENGTH parameter to specify the size of the buffer that Export uses for writing to the export file.
The RECORDLENGTH parameter specifies the length (in bytes) of the file record. You can use this parameter to specify the size of the Export I/O buffer (highest value is 64 kb). Changing the RECORDLENGTH parameter affects only the size of data that accumulates before writing to disk. It does not affect the operating system file block size. If you do not define this parameter, it defaults to your platform-dependent value for BUFSIZ (1024 bytes in most cases).
Then read metalink note: 155477.1
Or this one:
http://www.dba-oracle.com/oracle_tips_load_speed.htm
Then test in Your test env what are the best parameters for Your particular system.

BUFFER parameter in export

we know that BUFFER parameter in export
buffer_size = rows_in_array * maximum_row_size
but how/where can I find the rows_in_array and maximum_row_size
Please help me

To get row size
rem Table Sizes section
column Bytes heading BytesMB format 999,999,999,999
column tablespace_name heading 'Tablespace Name' format a16
column segment_name heading 'Table Name' format a35
column owner heading 'Owner' format a10
prompt ########################################################################
prompt
prompt Table Sizes Report
prompt
col tablespace_name format a20;
col segment_name format a30;
col owner format a8;
set linesiz 132;
set pagesiz 60;
BREAK ON owner ;
COMPUTE SUM OF bytes on owner;
select owner,tablespace_name,segment_name,sum((bytes/1024/1024)) Bytes
from sys.dba_extents
where owner in 'BAAN' and segment_type='TABLE'
group by tablespace_name,owner,segment_name
order by owner,tablespace_name,segment_name, bytes
http://www.baanboard.com/baanboard/showthread.php?t=13985

Will passing object as parameter increase memory usage?

Hi,
Will passing an object as a parameter increase the mmory usage? will it affect the performance of the application any means. Please share your view.
Regards,
A.

Ananth.durai wrote:
Hi ,
Just to precise my question, suppose i have an object called Student.I want to pass the Student Object to a method called addStudent().Now rather than passing Student Object, i am passing it's super class "Object". (i.e)
addStudent(Object object) instead of addStudent(Student student)will it make any difference in performance?Those are declarations, not method passing. You'd still be passing the same thing, if it's just a question about how you declare the method.
And it will make no difference in performance, except when you start getting bugs. You might start getting bugs because you declared the parameter with a more general type than is really necessary. If the method takes a Student, then declare it as taking a Student. Making it say it takes an Object will just make things worse.
But as per Cogniac,
when you pass a parameter or use an accessor function (get() function) or something of that sort, it's actually just passing a pointer that points to the memory location of the Object you want, not actually passing the Object itself to you, nor is it instantiating a new instance of the Object for your use.if i pass Object itself instead of Student,which memory address it will point out?That question makes no sense. You pass what you pass. If it's a Student object, then the reference the method gets will point to that object, regardless of whether the method knows that it's a Student or just an Object generally.

Change buffer parameter like rsdb/ntab/irdbsize

hi guruji
i wnat chang following buffer parameter in sap i try this from rz10 but some parameter i cant found for edit
i chagn following paremeter are i whant to change
ababp/buffersize,rsdb/ntab/irdbsize, etc
nainesh suthar

Hi Suthar,
The below mentioned parameters may not be defined in your profiles. It will be taking the default values, which can be checked using RZ11.
For editing these parameters, goto-->RZ11, select the profile and in extended maintenance, "create parameter" and specify the parameter with the required values.
Regards
Dona

Regd: Screen Buffer size increase

Dear Sir/madam,
We check in ST02. the screen Buffer -   SWAPS are More.
we want to increase the "SCREEN" Buffer size so the Screen Swaps comes down.
Screen buffer
zcsa/presentation_buffer_area    4400000          Byte   Size of screen buffer
sap/bufdir_entries               2000                    Max. number of buffered screens
Which parameter size i have to Increase here 2 Parameters are available. How much we can increase it.
here RAM size is 4GB. Any Formula is available to Increase the Buffer size of the Screen. or other like Program Buffer etc.
Regards
SNB.

Thanks for your Reply.
   Buffer            Hitratio   Allocated         Free space        Dir. size      Free directory         Swaps       Database
                        [%]       [kB]               [kB]         [%]      Entries         Entries       [%]                      accesses
Screen           96.99        4,297          141     3.44         2,000             1,722       86.10         1,506        4,525
if we want to increase we have to increase for both the parameters.
zcsa/presentation_buffer_area       4400000 + 440000 =          4840000 (10%)
sap/bufdir_entries                          2000 + 200 =                   2200   (10%)
What is the difference between thest 2 parameters.
1)                   zcsa/presentation_buffer_area
2)                   sap/bufdir_entries
Do you have any Formula to Handle ST02 Buffer Size Parameter Maintanence.
Regards
SNB.

Buffer parameter is better than direct=y

When I tried to export a history table that has 40 million rows & weighs 11GB, direct=y parameter is taking double the time compared to buffer=16000000 parameter.
exp uname/pwd@testdb file=/oradata/uat2.dmp buffer=16000000 statistics=none tables=(ALL_TXNS_MMDD)
buffer=16000000      *22 Minutes*
direct=y      *48 Minutes*
exporting onto local HDD
direct=y      *3 minutes*
buffer=16MB     *18 minutes*
Version- Oracle 9.2.0.4+
I always thought direct=y gives better speed and in all my export scripts I have used only direct=y.
does anyone know why buffer is giving better speed ?
Edited by: Sajiv on Jun 11, 2009 1:49 PM

Sajiv wrote:
When I tried to export a history table that has 40 million rows & weighs 11GB, direct=y parameter is taking double the time compared to buffer=16000000 parameter.
exp uname/pwd@testdb file=/oradata/uat2.dmp buffer=16000000 statistics=none tables=(ALL_TXNS_MMDD)
buffer=16000000      *22 Minutes*
direct=y      *48 Minutes*
exporting onto local HDD
direct=y      *3 minutes*
buffer=16MB     *18 minutes*
In your export command the @testdb suggests that you are exporting across a network. Does this mean that when you say "exporting onto a local HDD" you run the export command on the same machine as the database ?
If so then the difference in time may be related to the way in which SQL*Net can perform data compression for network traffic - and perhaps this can't be done in the same way for direct exports, leaving you with more round-trips when the traffic is across tcp/ip but fewer when using a local connection.
If you want to find out more about what's going on, you could simply query v$session_event for the session every few seconds to see where the time is going.
Regards
Jonathan Lewis
http://jonathanlewis.wordpress.com
http://www.jlcomp.demon.co.uk
"For every expert there is an equal and opposite expert."
Arthur C. Clarke

Exp buffer

Hi,
Version :10.2.0.1
What is buffer in export?
will it increase performance of exp?
I read the below link
http://docs.oracle.com/cd/B19306_01/server.102/b14215/exp_imp.htm
We have around 600 tables in a database? so How to calculate value for this parameter?
Thanks,

>
We have 600 tables in a database. I want full export.how to calculate appropriate value?
will it increase performance of exp?
>
Why are you ignoring the advice in the doc?
>
In general, Oracle recommends that you use the new Data Pump Export and Import utilities because they support all Oracle Database 10g features. Original Export and Import do not support all Oracle Database 10g features.
>
You should read the doc and follow the advice in it. That includes the advice to use expdp instead of export.
The BUFFER parameter applies only to conventional path Export.
Normally you would use the default buffer size; that is, do not specify BUFFER on the command line. The buffer size examples shown in the doc are specific to a table since they depend on the maximum_row_size and the number of rows that you want to fetch.

Export buffer maximum size

Hi,
For the parameter buffer used in export what is the maximum size we can give as input.

BUFFER
Default: operating system-dependent. See your Oracle operating system-specific documentation to determine the default value for this parameter.
Specifies the size, in bytes, of the buffer used to fetch rows. As a result, this parameter determines the maximum number of rows in an array fetched by Export. Use the following formula to calculate the buffer size:
buffer_size = rows_in_array * maximum_row_size
If you specify zero, the Export utility fetches only one row at a time.
Tables with columns of type LOBs, LONG, BFILE, REF, ROWID, LOGICAL ROWID, or DATE are fetched one row at a time.
Note:
The BUFFER parameter applies only to conventional path Export. It has no effect on a direct path Export. For direct path Exports, use the RECORDLENGTH parameter to specify the size of the buffer that Export uses for writing to the export file.
Example: Calculating Buffer Size
This section shows an example of how to calculate buffer size.
The following table is created:
CREATE TABLE sample (name varchar(30), weight number);
The maximum size of the name column is 30, plus 2 bytes for the indicator. The maximum size of the weight column is 22 (the size of the internal representation for Oracle numbers), plus 2 bytes for the indicator.
Therefore, the maximum row size is 56 (30+2+22+2).
To perform array operations for 100 rows, a buffer size of 5600 should be specified.
Ref. Oracle® Database Utilities
10g Release 2 (10.2)
Part Number B14215-01
Ch. 19 Original Export and Import
~ Madrid

From dng_image to pixel buffer?

Various things of the DNG SDK return a dng_image. The stage1, stage2 and stage3 are an RGB image. And if you use dng_render, you also get a dng_image.
I'd like to get a buffer of pixels (with for example R, G, B values). However, I haven't found any way in the API to get such a buffer.
For example, dng_image has member functions to get its width and height, but nothing to get the pixel buffer or even the value of a single pixel.
And I experimented a bit with the "Get" method of dng_image, but if I don't put anything in fArea of the buffer parameter, I didn't manage to get anything interesting, while if define a rectangle there, the whole program corrupts and crashes a bit after using dng_image.Get.
So, basically, my question is: how can I get the pixel buffer from a dng_image?
This is for displaying it on a computer monitor BTW.
Unfortunately the dng_validate sample doesn't help me out, because they don't get any pixel values there, they just call an API method to convert the dng_image to a TIFF file, which doesn't help me at all because I need RGB pixels on a screen, not a TIFF file on a disk.
Thanks.

I found the answer in the meantime, here it is: it's done with a dng_const_tile_buffer, where you give the dng_image as parameter in its constructur, and then you can get the pixels with dng_const_tile_buffer::ConstPixel.

Export and import of 3 Gb table

Dear ,
I have table with more than 3 million records, i need to export this table and import into a different database . can you please help me with the export and import options options so that i can speed up the process...
for you information
this table hase 3 Fk, 2 Indexs
i would appreciate your help in this regard
thanks

One way to speed up the exp/imp process is to use DIRECT=Y option. Also increase the size of BUFFER parameter.
While importing large amount of data it is better to drop relational constraints and indexes and recreate them after import.

Oracle 9.2 under Window2003 Release2

Hi colleague,
Always thanks yours good cooperation.
We are running server IBM system x3650 with 2 GB RAM and Opearating system Windows2003 Server Release 2 and we have
install oracle 9.2 but i am facing problem in increase there some parameter which are following.
Buffer Cache from 24 to 48
large pool from 08 to 504
when i increase them occured errors ORA-02097 and ORA-00384
Before Installation of oracle 9.2, we have define cache in page file
Initial Size : 3999
Maximum Size : 4094
Above paramer working prperly Oracle database 9.2 on windows2000 server but I am not understand about
this problem
Moreover, below parameter for your ready reference.
Shared_Pool_Size 48 MB
DB_Cache_Size 24 MB
Large_Pool_Size 8
Java_Pool_Size 32
SGA_MAX_SIZE = 129.069
Aggregate PGA Target 24 MB

you have asked this again and again..
Re: Oracle 9.2 Install on Operating System 2003
Re: Any body help me Urgengtly
SGA is the sum of the pools
In your case SGA_MAX_SIZE=129..
The sum of the pool size cannot be more than SGA_MAX_SIZE
Yours
[u]Shared_Pool_Size 48 MB
DB_Cache_Size 24 MB
Large_Pool_Size 8
Java_Pool_Size 32
SGA_MAX_SIZE = 129.069[/u]48+24+8+32=122
So 122 Mb which is less than 129MB(SGA_MAX_SIZE)---so you will be able to run
If you want to increase the buffer cache increase the SGA_MAX_SIZE first then increase you pool size or else you can utmost increase the size by only 6-7MB (129-122)whihc is left in the SGA_MAX_SIZE
Message was edited by:
Maran Viswarayar
Message was edited by:
Maran Viswarayar

How to speed up export import

Database 10.2.0
While doing export/import operation what are factors effecting the speed of operation..?

Below are the points which I collected from different links in my Oracle Notes:
Faster Export:
1. direct=y
2. subsets by query option
3. proper buffer size by rows in array * max row size
4. place the dump file to another disk (which is not having any datafile; to avoid I/O contention)
5. when database activity is low (in the morning or late eve)
6. PARALLEL
Faster Import:
1. indexes=n
2. constraints=n
3. user large buffer parameter
4. increase recordlength
5. commit=n
6. statistics=none
7. enough undo size
8. enough pga/sort_area_size
9. enough / Increase db_cache_size/db_block_buffers size and share_pool_size
10. when database activity is low (in the morning or late eve)
11. PARALLEL
Now check how your import is going on:
select
substr(sql_text,instr(sql_text,'into "'),30) table_name,
rows_processed, round((sysdate-to_date(first_load_time,'yyyy-mm-dd hh24:mi:ss'))*24*60,1) minutes,
trunc(rows_processed/((sysdate-to_date(first_load_time,'yyyy-mm-dd hh24:mi:ss'))*24*60)) rows_per_minute
from
sys.v_$sqlarea
where
sql_text like 'insert %into "%' and command_type = 2 and open_versions > 0;
Regards
Girish Sharma

HDV 1080p30 very slow when files are from Shake

Hi,
I have a totally tricked out 8 core Mac pro with 8 GB ram and a 1 TB drive (mostly empty). I'm using the Latest FCP. Recently, I've been editing some HD video that I've shot and have been quite pleased with the performance of the machine/software. However, I am experiencing a weird problem. I also have Shake, which I am using to do some composting and, relevantly, remove a small amount of camera vibration from my footage. (It works great.)
However, any files that I export from Shake, using (as best as I can tell) the exact same file formats and codecs that my imported Raw (unprocessed in Shake) video is. That is: they are quicktime files, Using HDV 1080p30fps, with 29.97 FPS "drop frame" codec. While the raw imported files work great, the files I export from Shake and put into my timeline result in a "Spinning Ball" whenever I do anything at all to them- play them, move the timeline around, cut, apply a filter etc,. What am I overlooking?
Brad Hoehne

Hi Kunal
Let me explain to you how this works
When you archive the data, it moves out of the database into a flat file which is latter stored on optical disk/hard disk etc depending upon your archiving solution
When archving is done, it also creates an index ( depending upon how your have configured the object) This index is of few common fields of the archived objects . This is than stored within SAP. When you call SARI you seen this indexed data from SAP and hence the response is quick. At this point it does not get the data from your archives
When you call it through the transaction FBL.. , it needs all the fields ( which are not avl in the indexed data ) and hence SAP has to go to the archives to fetch it
This results in slowing it down.
Now how slow will it be is a difficult question for us to answer as a lot denpends upon your archiving solution and how much data are you trying to fetch
As far as memory problem is concerned you need to analsye the dump as to where ( on what buffer parameter is it failing) Depending upon that you will have to either increase memory parmeters OR reduce the amount of data your a fetching
Thanks
Naushad

Export take long hrs

Hi
I export 38 gb db size , but it takes 10 hrs , last time i use datapump export
that one is taking long hrs comparing to exp utility
now exp also taking long hrs , table contains lobs , how i can tune this export
this export is running in oracle 10g version
rds

HI
I read this note from oracle web site that is why i didnot increase the buffer size,
regarding the LOB with direct =y will not support in 8 & 9 version , but i did'nt
encounter any problem so for in 10g , but if you got any information regarding this
lob expotr in 10g with Direct=y , pls let me know
quote from oracle
The BUFFER parameter applies ONLY to conventional path Export. It has no
effect on a direct path Export.
This BUFFER parameter specifies the size (in bytes) of the buffer used to
fetch rows. It determines the maximum number of rows in an array, fetched
by Export.
For direct path Export, use the RECORDLENGTH parameter to specify the size
of the buffer that Export uses for writing to the export file.
Note that when you use Export release 8.0 on an Oracle9i database, then
Export doesn't export rows from tables containing objects and LOBs when
you have specified a direct path load (DIRECT=y).
In my export log i can see the row containg lobs are succesfully exported
--. . exporting table BIZDOCCONTENT 386222 rows exported
This means it is working corrcet
pls let me know
Best rds

Buffer Parameter Increase???

Similar Messages

Maybe you are looking for