Cpu usage high when loading cache group

Hi,
What are the possible reasons that results high cpu usage when loading read-only cache group with big root table (~ 1 million records)? I have tried setting Logging=0 (without cache agent), 1 or 2 but it doesn't help. Are there any other tuning configuration required to avoid high cpu consumption?
ttVersion: TimesTen Release 6.0.2 (32 bit Solaris)
Any help would be highly appreciated. Thanks in advance.

High CPU usage is not necessarily a problem as long as the CPU is being used to do useful work. In that case high CPU usage shows that things are being processed taling maximum advantage of available CPU power. The single most common mistake is to not properly size the primary key hash index in TimesTen. Whenever you create a table with a PK in TimesTen (whetehr it is part of cache group or just a standalone table) myou must always specify the size of the PK hash index using the UNIQUE HASH ON (pk colukns) PAGES = n clause (see the documentation). n should be set to the maximum number of rows expected in the table / 256. The default is sized for a table of just 4000 rows! If you try and load 1M rows into this table we will be wasting a lot of CPU time serially scanning the (very long) hash chains in each bucket for every row inserted...

Similar Messages

CPU Usage high when USB eneabled

I have a PC running Windows XP with a MS6340 (V5) motherboard and Athlon XP 1800 processor.
When getting a poor response I noticed that the performance tab on Task manager showed that CPU usage was wildly fluctuating between 50-60%. This was farily soon after booting up and nothing else was running. The Processes tab on task manager showed that there were no processes running which got anywhere near accounting for this.
I downloaded a program called Iarsn TaskInfo2002 which shows that the CPU Usage was because of something called Interrupts Time placeholder and (to a lesser extent) DPC Time Placeholder.
Anyway the interesting thing is that by disabling my USB ports (via device manager) and then rebooting the problem was gone. (Interrupts Time and DPC Placeholders don't clock anything.)
Now, however, I have a USB camera and Printer and so MUST get my USB ports enabled again.
I can enable them after booting and then the problem is not as bad, but I nearly always forget to disable them again before shutting down and then next time its a pain.
Is this likely to be
A motherboard problem,
BIOS problem
Windows XP problem
or me being stupid.
Any (polite) advice would be gratefuilly received
SIDDY

Hi,
It's not a motherboard problem.
As the board only provides the ports, nothing more.
The program that queries the ports for data is most likely the problem.
Try deinstalling the printer or camera software, see what happens.

HP Load Runner: CPU usage for this load generator has exceeded 80%

Recently I have downloaded LR 12.1 and installed. Observed error CPU usage for this load generator has exceeded 80% when executing the test with 5 user load using AJAX Trulcient IE script. For each virtual user in the script one TcWebIELauncher.exe hads been initiated and running during the test execution. Each TcWebIELauncher.exe instance is taking min 14% CPU utilization and CPU utilization for 5 users is around 75% at the time of test execution. What is the reason behind this high CPU utilization for this less user load?. Please suggest how to handle and fix the issue.
regards,
Sreekanth

You can download Trail version of Loadrunner from HP site. That's true. all HP testing tools products are license based.
i) If you want to use for SAP applications, you need to have SAP protocals (SAP GUI and SAP WEB). So how this licenses works is, if you want to buy 50 Vusers for SAP it cost you around $ 32k and it depends on company discount. so you need to work with HP Sales team to get the quotation then based on no of Vusers and you can procure it.
ii) As for as i know, there is no HP LR integration with Solution Manager.
let me know if you have any questions.

Error loading Cache group but Cache group created with out error

Hi
I have created a cache group but when I load that cache group I get following error:
Command> load cache group SecondCache commit every 1 rows;
5056: The cache operation fails: error_type=<Oracle Error>, error_code=<972>, error_message:ORA-00972: identifier is too long
5037: An error occurred while load TESTUSER.SECONDCACHE:Load failed (ORA-00972: identifier too long)
The command failed.
Please help.
Looking forward for your reply.
/Ahmad

Hi Chris!
Thanks for urgent response. I solved my problem to some extent but want to share.
Acctualy I was having a column named # which is a primary key also. When I change that column name from # to some other name like some characters then the cahe group is loaded successfuly.
Is there anyway in TimesTen to load columns names # .
I read in the documentation of TimesTen that it allows columns names as # , so it is the reason it is creating the cache group but fails to load do not know the reason.
The code for creating cache group is as follows:
create cache group MEASCache from testuser."MEAS"("UPDATED" number not
null,"UNOCCUPIEDRECORD" number not null,"VALUECURRENT" number not null,"EQSFREF
" number not null,"IMPLEMENTED" number not null,"FORMAT" number not null,"#" number not null,primary key("#"))
When I change the # column to like eg Identity it works fine.
/Ahmad

Load cache group with parallel error, 907

hello, chris:
we met another question, when we create a cache group, then load the data with parallel 8, it appeared unique conflict, we check the data but didn't found any data question, so we load the data again without parallel parameter, it works well, all the data load in. then use unload and load with parallel 8 again, it appeared unique confict again, what happend??
thank you...
The script ls:
create readonly cache group FSZW_OCECS.SP_SUBSCRIBER_RELATION_CACHE
autorefresh
mode incremental
interval 29000 milliseconds
/* state on */
from
FSZW_OCECS.SP_SUBSCRIBER_RELATION (
SUBS_RELATION_ID TT_BIGINT NOT NULL,
PRIVID VARCHAR2(32 BYTE) INLINE NOT NULL,
SUBSID TT_BIGINT,
SWITCH_FLAG VARCHAR2(2 BYTE) INLINE,
DISCOUNT_CODE VARCHAR2(8 BYTE) INLINE NOT NULL,
DISCOUNT_SERIAL TT_INTEGER,
START_DATE DATE NOT NULL,
END_DATE DATE,
MOBILENO VARCHAR2(15 BYTE) INLINE NOT NULL,
APPLY_DATE DATE,
primary key (SUBS_RELATION_ID))
where NODEID = '334' or NODEID IS NULL,
FSZW_OCECS.SP_SUBSCRIBER_ATTRINFO (
SUB_ATTACH_ID TT_BIGINT NOT NULL,
SUBS_RELATION_ID TT_BIGINT,
SUB_ATTACH_INFO VARCHAR2(16 BYTE) INLINE NOT NULL,
SUB_ATTACH_TYPE VARCHAR2(2 BYTE) INLINE,
primary key (SUB_ATTACH_ID),
foreign key (SUBS_RELATION_ID)
references FSZW_OCECS.SP_SUBSCRIBER_RELATION (SUBS_RELATION_ID));
Command> load cache group SP_SUBSCRIBER_RELATION_CACHE commit every 25600 rows PARALLEL 8;
5056: The cache operation fails: error_type=<TimesTen Error>, error_code=<907>, error_message: [TimesTen]TT0907: Unique constraint (SP_SUBSCRIBER_ATTRINFO) violated at Rowid <0x0000000091341e88>
5037: An error occurred while loading FSZW_OCECS.SP_SUBSCRIBER_RELATION_CACHE:Load failed ([TimesTen]TT0907: Unique constraint (SP_SUBSCRIBER_ATTRINFO) violated at Rowid <0x0000000091341e88>
Command> load cache group FSZW_OCECS.SP_SUBSCRIBER_RELATION_CACHE commit every 25600 rows;
5746074 cache instances affected.

This looks like a bug to me but I haven't been able to find a known candidate. Are you able to log an SR and provide a testcase so we can reproduce it here and verify if it is a new bug? Thanks.

How to load cache group?

Dear ChrisJenkins,
My project has a timesten . There is a table (using read only cache group) in timesten.
ex :
create table A as (id number, content varchar(20));
insert into A values (1, 'a1');
insert into A values (2, 'a2');
insert into A values (n, 'an');
commit;
The table (A) loaded 10 rows ('a1' --> 'a10'). if I execute the sql following :
"Load cache group A where id >=2 and id <=11"
, how will the timesten execute the sql above ?
I suggest :
the timesten don't load rows (id=2-->10) because the memory has the rows ,
the timesten only load rows (id=11) because the memory don't has the row
Is it true ?
Thanks,rgds
TuanTA

In your example you are using a regular table not a readonly cache group table. If you are using a readonly cache group then the table would be created like this:
CREATE READONLY CACHE GROUP CG_A
AUTOREFRESH MODE INCREMENTAL INTERVAL 10 SECONDS STATE PAUSED
FROM
ORACLEOWNER.A ( ID NUMBER, CONTENT VARCHAR(20));
This assumes that the table ORACLEOWNER.A already exists in Oracle with the same schema. The table in TimesTen will start off empty. ALso, you cannot insert, delete or update the rows in this table directly in TimesTen (that is why it is called a READONLY caceh group); if you try you will get an error. All data for this table has to originate in Oracle. Let's say that in Oracle you now do the following:
insert into A values (1, 'a1');
insert into A values (2, 'a2');
insert into A values (10, 'a10');
commit;
Stilll the table in TimesTen is empty. We can load the table with the data from Oracle using:
LOAD CACHE GROUP CG_A COMMIT EVERY 256 ROWS;
Mow the table in TimesTen had the same rows as the table in Oracle. Also, the LOAD operation changes the AUTOREFRESH state from PAUSED to ON. You still cannot directly insert/update and delete to this table in TimesTen but any data changes arising due to DML executed on the Oracle table will be captured and propagates to TimesTen by the AUTOREFRESH mechanism. If you now did, in Oracle:
UPDATE A SET CONTENT = 'NEW' WHERE ID = 3;
INSERT INTO A VALUES (11, 'a11');
COMMIT;
Then, after the next autorefresh cycle (every 10 seconds in this example), the table in TimesTen would contain:
1, 'a1'
2, 'a2'
3, 'NEW'
4, 'a4'
5, 'a5'
6, 'a6'
7, 'a7'
8, 'a8'
9, 'a9'
10, 'a10'
11, 'a11'
So, your question cannot apply for READONLY cache groups...
If you used a USERMANAGED cache group then your question could apply (as long as the cache group was not using AUTOREFRESH and the table had not been marked READONLY). In that case a LOAD CACHE GROUP cpmmand will only load qualifying rows that do not already exist in the cache table in TimesTen. If rows with the same primary key exist in Oracle they are not loaded, even if the other columns have different values to those in TimesTen. Contrast this with REFRESH CACHE GROUP which will replace all matching rows in TimesTen with the rows from Oracle.
Chris

CPU/Memory Usage increases when loading SWF files in AIR.

Hi,
We face the below mentioned problem. Would be great if someone could help us.
Due to problems with uploading the attachment here, please send a email to the following email-id ([email protected]) and I shall send the attachment in response.
CPU/Memory usage increases drastically while moving the SWC components and loading the SWF Files.
"Problem Description" - On executing the sample application, a window with multiple objects is shown. The loaded SWC components in flex consume too much Memory and CPU cycles. This is visible in TaskManager. On dragging one of the object, CPU / memory usage increases drastically.
"Steps to Reproduce" -
Draggable window is placed at the bottom in the sample application with Text indication.
Drag the indicated window.
"Actual Result" - System's CPU usage increasing drastically around 50% to 60 %.
"Expected Result" - Drastical increase in memory\CPU should not occur.
"Any Workarounds" - No"
Thanks in advance.

You can control your external swf from your main.swf in the following method:
loadswf(clips[0]);
function loadswf(tmp:String):void{
    passedString=tmp;
    ldr= new Loader();
    ldr.contentLoaderInfo.addEventListener(ProgressEvent.PROGRESS, progressListener);
    ldr.contentLoaderInfo.addEventListener(Event.COMPLETE, swfLoaded);
    ldr.load(new URLRequest(animPath+tmp));
    function progressListener (e:ProgressEvent):void{
        ploader.visible=true;
    function swfLoaded(e:Event):void {
        mcExt = e.target.content as MovieClip;
        ldr.contentLoaderInfo.removeEventListener(Event.COMPLETE, swfLoaded);
        mcExt.addEventListener(Event.ENTER_FRAME,onEnterfn);
        ploader.visible=false;
        addChild(mcExt);
function onEnterfn(e:Event):void{
    var num:int=mcExt.currentFrame;
    switch(passedString){
        case clips[0]:
            if (num==1) mcExt.play();
            if (num==9) {
               mcExt.skipmc.addEventListener(MouseEvent.CLICK,skipfn);
                function skipfn(e:MouseEvent):void{
                    mcExt.stop();
                   mcExt.removeEventListener(Event.ENTER_FRAME,onEnterfn);
                    gotoAndStop("help");
            if (num==mcExt.totalFrames){
               mcExt.removeEventListener(Event.ENTER_FRAME,onEnterfn);
                gotoAndStop("help");
            break;

What Causes High CPU Usage Even When PC is Idle

How can I fixed the high cpu usage on my pc?

mcmillan wrote:
rohankolay wrote:Right now i have kernel26-2.6.35.4 which i updated on 27 aug. That is around the same time the problem began.Should i downgrade it to kernel26-2.6.35.3 or kernel26-2.6.34.3?
I think most people have been seeing the problems in any of the 2.6.35 kernels, so 2.6.34.3 would be my recommendation
+1
And put kernel26 in IgnorePkg.

MBP with 10.6.8: CPU usage high, no responsible processes shown in Activity Monitor. What causes this?

After using my computer for some time, CPU usage climbs to about 90% (ca. 65% user / 26% system) and fans run at high speed, but there is no processes that seem to use CPU (other than Activity Monitor itself ~1%).
This has happened a few times recently. It happens when I've used my computer for some time, hours or a couple of days (I usually put my computer to sleep when not using it but rarely shut it down). I've tried to quit all running apps incl. Growl, Dropbox, and anything that could be responsible to this but with no results. Putting computer to sleep for a long time doesn't help, cpu usage will climb instantly after wake up. Restarting system is the only thing that helps.
Most of the time I use my laptop with external monitor, lid closed and connected to power supply. When experiencing this, opening the lid causes fans to slow down, but CPU usage remains high and battery drains quick.
Any ideas what might cause this?
My MBP is 15" model with Core 2 Duo 2,66 GHz (MacBookPro 5,3), OS X version is 10.6.8, all available updates are installed.

Perform a SMU reset.
https://support.apple.com/kb/HT3964
Remove the anti-virus, it's running a root level process all the time that's taking your CPU and it's not necessary as there are no Mac viruses, Apple has the trojan angle covered already.
Always on anti-virus on a Mac is always a source of problems, also is MacKeeper, uninstall that here if you have it.
http://applehelpwriter.com/2011/09/21/how-to-uninstall-mackeeper-malware/
If you need to clean the Windows filth off PC files, then use the free, run as you need, ClamXav.
http://www.clamxav.com/
It seems your machine is used in a corporate enviroment, thus likely a rule that anti-virus must be used, you can avoid high CPU usuage by using the ClamXav instead.

CPU usage 100% when WLS 10.3.3 is idle on OS X 10.6.4 (Snow Leopard)

I've installed WLS 10.3.3 on OS X 10.6.4 and executed using the instructions in the README (including setting the MEM_ARGS).
Upon startup, the CPU usage spikes to ( then stays ) around *104%* ( dual-core CPU ); this, of course cripples the computer to where all running applications suffer. Attempting to access the console, brings the system to it's knees. This happens even if it is the only (non-system) application running.
We need to be able to run WLS on our local development machines (MacBook Pros from 2007-present).
We are currently running OAS (OC4J) on the machines with out any issues (CPU usage at OAS idle = ~1.1%).
Are there any WLS performance settings that can be adjusted to alleviate the CPU utilization issue?
My System Info:
Model Name:     MacBook Pro
Model Identifier:     MacBookPro4,1
Processor Name:     Intel Core 2 Duo
Processor Speed:     2.4 GHz
Number Of Processors:     1
Total Number Of Cores:     2
L2 Cache:     3 MB
Memory:     4 GB
Edited by: user8664206 on Aug 3, 2010 8:54 AM

Thank you!
The key was modifying the "setDomainEnv.sh" file (this file is generated using the domain GUI configuration tool).
I added the following at the top of the script (just below the comments):
JAVA_VENDOR="Sun"
export JAVA_VENDOR
I modified:
SUN_JAVA_HOME=""
export SUN_JAVA_HOME
...to be:
SUN_JAVA_HOME="/System/Library/Frameworks/JavaVM.framework/Versions/1.6.0/Home"
export SUN_JAVA_HOME
Now when I execute the container, it consumes only 0.2% of the CPU.
Thanks again,
Jim

IPhoto cpu usage high, process isboxd

Hi All
Hoping you could shed some light on an issue i'm having with iPhoto. I've recently noticed that on launch it has been running a process called isboxd, and this is using a high amount of cpu power, and also pushes my core temp up to 80 degrees C. Once i quit iPhoto this process stops and cpu usage drops back to its normal 1%. I've tried quitting the process whilst iPhoto is running and it immediately restarts itself. I have also rebuilt my iPhoto library too and this hasn't sorted the issue either.
My iPhoto library is 56GB
I've only just started noticing this process happening, hopefully someone can shed some light onto what the process is doing, and if i can stop it.

Then try to test in a different user account. You might want to enable the guest account for testing. If you are seeing the same proble in a different account, that would show that your iPhoto preferences may be corrupted.
Delete the iPhoto preference file: com.apple.iPhoto.plist from your ~/Library/Preferences folder.
In Lion and Mt. Lion also the cached file in
~/Library/Containers/com.apple.iPhoto/Data/Library/Caches/com.apple.iPhoto
The library is hidden in Mt. Lion: To remove the files:
Click somewhere on the empty Desktop to bring the Finder forward.
Select "Go" in the Finder's main menu bar and hold down the alt/option â¥ key, until you see the "Library" in the drop-down menu. Double click it to open a Finder window, so you can browse the files inside.
Scroll down to the "Preferences" folder and open it.
Remove com.apple.iPhoto.plist
Note: this will reset all iPhoto preference settings to the default values, you may want to take a note of them, before you do that.

CPU usage High

Hi,
Can any one tell me when the CPU usage will go high in oracle? What are the various reasons that affect CPU usage? what are the necessary steps we need to take in order to bring down?
I will be grateful to your answer
With Regards
Boo

CPU usage will increase in Oracle when what you are doing requires CPU resources.
To find of what those things are read the docs.
The steps one uses to decrease CPU utilization depend on what is causing it.
If you want to read the docs you will find them at http://tahiti.oracle.com
If you want someone else to read them for you hire a tutor.

CPU Usage High, Individual Apps Not using Total

Hi everyone,
I've got a 2007 white Macbook currently running the most up to date version of Snow Leopard and recently the fan has been on a lot more often than it usually is and when I check in Acitivity Monitor the CPU usage is extremely high. Normally it's around 2-4% tops but several times over the last few months it's been at 50-60% sustained usage for a long time (upwards of 20 minutes, then I just reboot it). Once or twice it waws higher than 100%. I checked Activity Monitor and killed the normal culprits like Shockwave and Flash, then anything with Microsoft in the name. I'm a big tabbed browser user so I closed off google Chrome. Right now, running Adium, iCal, Powerpoint (small file size, only 5 slides), and Word, 45% of the CPU is currently in use. I went through and systematically closed everything but the CPU usage is still around 45% after closing all active programs in the dock other than Finder and Activity Monitor. The strange part for me is that the amount of CPU usage is not the sum of the CPU usage of all the programs currently running. It's not even close (like 1% versus 45%)
Does anyone know what might be wrong? Rebooting it will usually fix this problem, but I'm worried I either have Malware, a Trojan that iAntivirus isn't detecting, and my computer is somehow being used without my knowledge or something is wrong with my Mac.
Thanks in advance for any help!
Specs are Macbook 3,1, Intel Core 2 Duo 2.2 GHz, 4GB ram

hrishy wrote:
Hi
Why so much time is being spent in CPU ?
Apologies the rich text editor is not working and i am unabel to format this code properly
Rows Operation
______0 NESTED LOOPS (cr=117469550 pr=444 pw=0 time=1670966102 us)
(snip)
______0                    NESTED LOOPS (cr=117469550 pr=444 pw=0 time=1670965844 us)
462,472                       NESTED LOOPS (cr=1389078 pr=196 pw=0 time=12429546 us)
(snip)
______0                       TABLE ACCESS FULL CON (cr=116080472 pr=248 pw=0 time=1889064945 us)
(snip)
The execution plan shows a nested loop join between a row source containing 462,472 rows and a full table scan of the table CON - that means the table CON was full table scanned 462,472 times, resulting in 116,080,472 consistent gets in 9 executions. The consistent gets are the cause of the high CPU usage. You might work through the links provided by Narendra to find the cause of the nested loop joins. Maybe this indicates a problem with the object statistics, or possibly it is caused by an unfortunate set of parameters that affected the optimizer's decisions.
Charles Hooper
Co-author of "Expert Oracle Practices: Oracle Database Administration from the Oak Table"
http://hoopercharles.wordpress.com/
IT Manager/Oracle DBA
K&M Machine-Fabricating, Inc.

Satellite R830-1GZ - CPU usage high ( 25%) and fan speed increase after wakeup

Hello all
The CPU usage of my laptop shoots up to 25% (and beyond) and the fan speed increases dramatically after waking up, but only when I'm using an external mouse (Microsoft Natural Wireless Laser Mouse 6000) - disconnecting the mouse once this has happened does not solve the problem, so I'm always forced to restart my computer if I have let it go to sleep with the mouse connected.
I've noticed that when this happens System Interrupts - Deferred Procedure Calls and Interrupt Service Routines appears in the Resource monitor>CPU>Processes, and seems to explain the increase in CPU usage.
Any ideas on how to fix this would be most appreciated.
I'm using a Toshiba Satellite R830-1GZ with Windows 7 (64 bit). This problem does not happen when I use the same mouse with another laptop (Dell with Windows XP).
Thank you very much.
CM

Thank you for your reply.
I was using the mouse without installing the software that came with it (IntelliPoint), which seems to have been necessary to obtain the latest driver. So I've installed the latest driver now, but... I've still come across the same high CPU usage problem after wakeup.
If your suspicion is correct, should disconnecting the mouse before putting the computer to sleep (hopefully allowing the process that controls the mouse to close properly) stop this problem from occurring when it wakes up?
Could this problem also occur due to an external keyboard? I'm pretty sure it's the mouse (given my experience of using only the external mouse and only the external keyboard) and my understanding is that keyboards don't require special drivers (at least the A4Tech one I have claims it doesn't)... just trying to exclude all possibilities.
Thanks again for your help.

3.2 - CPU usage high

Ocaasionally SQL Developer comes in a state where CPU usage is 50% (on a dual core machine) and according to jvisualvm "Background Parser" and InsightThread is using the CPU, and only these two.
This continues also after closing all connections.
I don't really know what lead up to this.
Version 3.2.09/Build MAIN-09.23
Edited by: jnp1234 on Aug 28, 2012 7:01 AM

Hi,
This temporary high CPU usage is most likely related to the size of your SQL History. Look for the preference that controls the size limit in Tools | Preferences | Database | Worksheet.
Regards,
Gary
SQL Developer Team

Cpu usage high when loading cache group

Similar Messages

Maybe you are looking for