Performance change after upgrade

I was able to execute a procedure in Oracle 8: 8.0.4 in under 10 seconds that now takes over 2000 seconds with our server being upgraded to 8i: 8.1.7. The procedure involves a number if insert statements involving views as the underlying source code. The time comparisons involve one record being parsed for each of the insert statements. The only change has been the upgrade to 8i. Any ideas as to why this has occurred? This seems to point to the procedure running outside the memory cache which may be too small now. Our contractor DBA is not helping at this time. Any suggestions??

Analyze your tables, indexes and indexed columns and see if this makes a difference.
e.g.,
Analyze table my_table compute statistics
FOR TABLE
FOR ALL INDEXES
FOR ALL INDEXED COLUMNS ;

Similar Messages

  • When creating a new window in safari, using command T, the marker have not activated the search field. It means you have to click there every time. Changed after upgrading to Mavericks. Anyone knows how to change this?

    When creating a new window in safari, using command T, the marker have not activated the search field. It means you have to click there every time. Changed after upgrading to Mavericks. Anyone knows how to change this?

    Yes, you can do something like that. What you would do is create a button for each image and then hide them. When I do this type of thing I place all of the buttons on a template page and then hide the template.
    You can then use JavaScript to copy the icon from any of the hidden buttons to the main button that you've set up to display the image. For example, suppose you set up 10 buttons named b1, b2, b3, ...b10. The code to copy the icon from one of them to the button used to display the image (b0) would be something like this:
    // Get the icon from the b3 button
    var oIcon = getField("b3").buttonGetIcon();
    // Get  reference to the b0 button
    var f = getField("b0");
    // Set the icon of the b0 button to the icon retrieved from the hidden button
    f.buttonSetIcon(oIcon);
    // Show the b0 button
    f.display = display.visible;
    This code would go in the Mouse Up event of the smaller button.

  • Performance issue after Upgrade from 4.7 to ECC 6.0 with a select query

    Hi All,
    There is a Performance issue after Upgrade from 4.7 to ECC 6.0 with a select query in a report painter.
    This query is working fine when executed in 4.7 system where as it is running for more time in ECC6.0.
    Select query is on the table COSP.
    SELECT (FIELD_LIST)
            INTO CORRESPONDING FIELDS OF TABLE I_COSP PACKAGE SIZE 1000
            FROM  COSP CLIENT SPECIFIED
            WHERE GJAHR IN SELR_GJAHR
              AND KSTAR IN SELR_KSTAR
              AND LEDNR EQ '00'
              AND OBJNR IN SELR_OBJNR
              AND PERBL IN SELR_PERBL
              AND VERSN IN SELR_VERSN
              AND WRTTP IN SELR_WRTTP
              AND MANDT IN MANDTTAB
            GROUP BY (GROUP_LIST).
       LOOP AT I_COSP      .
         COSP                           = I_COSP      .
         PERFORM PCOSP       USING I_COSP-_COUNTER.
         CLEAR: $RWTAB, COSP                          .
         CLEAR CCR1S                         .
       ENDLOOP.
    ENDSELECT.
    I have checked with the table indexes, they were same as in 4.7 system.
    What can be the reson for the difference in execution time. How can this be reduced without adjusting the select query.
    Thanks in advance for the responses.
    Regards,
    Dedeepya.

    Hi,
    ohhhhh....... lots of problems in select query......this is not the way you should write it.
    Some generic comments:
    1. never use SELECT
                       endselect.
       SELECT
      into table
       for all entries in table
      where.
       use perform statment after this selection.
    2. Do not use into corresponding fields. use exact structure type.
    3. use proper sequence of fields in the where condition so that it helps table go according to indexes.
        e.g in your case
              sequence should be
    LEDNR
    OBJNR
    GJAHR
    WRTTP
    VERSN
    KSTAR
    HRKFT
    VRGNG
    VBUND
    PARGB
    BEKNZ
    TWAER
    PERBL
    sequence should be same as defined in table.
    Always keep select query as simple as possible and perform all other calculations etc. afterwords.
    I hope it helps.
    Regards,
    Pranaya

  • Performance issues after upgrading to 11g

    hello,
    We are facing a performance issue after upgrading from 10g to 11g.
    as you can see, the tables named APPLSYS.FND_ENV_CONTEXT has grown to 15544880 records,
    and oracle is using TABLE ACCESS FULL on it:
    DELETE FND_ENV_CONTEXT E
    WHERE NOT EXISTS
    (SELECT 'X'
    FROM FND_CONCURRENT_PROCESSES P
    WHERE P.CONCURRENT_PROCESS_ID = E.CONCURRENT_PROCESS_ID)
    AND ROWNUM < 10000;
    Plan:
    DELETE STATEMENT ALL_ROWSCost: 74,345                     
         5 DELETE APPLSYS.FND_ENV_CONTEXT                
              4 COUNT STOPKEY           
                   3 HASH JOIN RIGHT ANTI Cost: 74,345 Bytes: 3,108,980 Cardinality: 155,449      
                        1 INDEX FULL SCAN INDEX (UNIQUE) APPLSYS.FND_CONCURRENT_PROCESSES_U1 Cost: 166 Bytes: 315,450 Cardinality: 63,090
                        2 TABLE ACCESS FULL TABLE APPLSYS.FND_ENV_CONTEXT Cost: 69,124 Bytes: 233,173,200 Cardinality: 15,544,880
    select num_rows,last_analyzed from dba_tables where table_name='FND_ENV_CONTEXT'
    15544880 07/07/2011 12:08:55
    please advice.
    Ofer

    Please see these docs.
    Table Fnd_env_context Growing Very Fast [ID 419990.1]
    FND Related Tablespaces Growing at Rapid and Excessive Rate [ID 189800.1]
    FNDCPPUR Request Does Not Always Delete Files From The File System, Many Files Will Be Denoted As "deletion failed" [ID 1335304.1]
    Concurrent Processing - Best Practices for Performance for Concurrent Managers in E-Business Suite [ID 1057802.1]
    Concurrent Processing - Purge Concurrent Request and/or Manager Data Program (FNDCPPUR) [ID 104282.1]
    Thanks,
    Hussein

  • Apple logo will be changed after upgrade iphone5

    i have iphone 5.will apple logo be changed after upgrade?

    I had this same thing happen to me.  I was using my brand new iphone 5 (texting and listening to music through the speakers) when it flashed the apple logo, stopped my music and brought me to the lock screen. 

  • Performance degradation after upgrading to yosemite

    I'm experiencing performance degradation on MacBook Pro 15  including
    number of spinning wheels,
    instance of dark screen, 
    overheating after upgrading to Yosemite
    diminished battery life
    Is Yosemite the cause of this and other issues

    I can't believe Apple have made this software available to the masses when it's clearly not ready. Since "upgrading" my Mac mini has been pretty useless. Programs take a substantially increased time to open compared to Mavericks, and then performance is poor at best. Opening files in Photoshop, for example, is painfully slow and I get the colour wheel almost all of the time - something which definitely didn't happen with Mavericks.
    Aside from this, I've experienced a number of frustrating bugs today. When I open a new program on my left screen the right screen is changed to a fresh desktop. Why? When I turn my Mac on I get the login screen on the left monitor (as it always used to) but then the primary desktop is set to the right and I've been unable to keep the correct setting so far (it forgets that I've changed it following a reboot). The background of the top bar keeps disappearing so all of the icons, time, etc. just sit on top of the desktop background.
    I hope Apple can release a fix, and quickly.

  • The optimizer changed after upgrade from 8i to 10g

    I have upgraded my Oracle database from 8i (8.1.7.3) to 10g (10.2.0.3). After upgraded, there's some SQLs are with performance issue that the response time is very slow. The SQL can run very fast (around 1 seconde) in the pre-upgrade environment. Also, after generating the execution plan, there's different from the original enivronment (8i) which the query using some full table scan instead of index scan originally used in 8i.
    The optimizer_mode of 8i is RULE and it's CHOOSE when upgraded.
    My questions are:
    1. Do I have any init.ora parameters setting in 10g environment which differ from 8i?
    2. What method of analyzing tables is appropriate to use in 10g? Will this affect the performance like my case?
    Thanks,
    Wai

    Please have a look to following threads:
    Re: Optimizer Statistics collection after upgrade from 8i to 10R2
    Re: DB upgrade from oracle 8.1.7 to oracle 10.2.0.2

  • Report Layout Changed After Upgrade from 4.6C to ECC6

    Dear Gurus,
    I do functional test of every process in ECC6 after upgrade from 4.6C. I found all report layout changed without variant (it's happen to all Query/SQ01 reports ). So I've to assign manually the fields for the reports, and create a default variant then.
    Why it happen ? Is there configuration or setting missed or this is standard while we do upgrade ?
    Thx in advance.
    Tri Wicaksono

    We do re arrange for all ALV report ...

  • SharePoint 2007 performance issue after upgrading the operating system

    We upgraded the operating system of our SharePoint Server 2007 from 2003 to 2008 R2. After upgrade we rebuilt the SharePoint using content DBs we backed up before upgrade. All settings is same except that we moved the DBs to new SQL servers. Now users are
    reporting performance issue with sites. I experienced it myself some of the pages takes 9-30 seconds to load. Also our SharePoint server is single server. I monitored the SharePoint Server and it is fine. I did fiddler trace and here are the result. I see
    lots of 401 (login failed) errors then 302 (redirect) then 200 (OK). Do you know why we are getting 401 errors?
    Result
    Protocol
    Host
    URL
    Body
    Caching
    Content-Type
    Process
    184
    401
    HTTP
    sp07.ourcompany.com
    /bif/EUROPE-AMA
    341
    text/html; charset=us-ascii
    iexplore:15344
    185
    302
    HTTP
    sp07.ourcompany.com
    /bif/EUROPE-AMA
    196
    text/html; charset=UTF-8
    iexplore:15344
    186
    401
    HTTP
    sp07.ourcompany.com
    /bif/EUROPE-AMA/default.aspx
    341
    text/html; charset=us-ascii
    iexplore:15344
    187
    200
    HTTP
    sp07.ourcompany.com
    /bif/EUROPE-AMA/default.aspx
    105,487
    private, max-age=0; Expires: Mon, 22 Dec 2014 21:37:48 GMT
    text/html; charset=utf-8
    iexplore:15344
    188
    401
    HTTP
    sp07.ourcompany.com
    /_layouts/1033/styles/calendar.css?rev=BrbrIU86qTG2EHx1ZUuFBQ%3D%3D
    341
    text/html; charset=us-ascii
    iexplore:15344
    189
    200
    HTTP
    sp07.ourcompany.com
    /_layouts/images/prevbuttonltr.gif
    76
    max-age=31536000
    image/gif
    iexplore:15344
    190
    200
    HTTP
    sp07.ourcompany.com
    /_layouts/images/nextbuttonltr.gif
    78
    max-age=31536000
    image/gif
    iexplore:15344
    191
    200
    HTTP
    sp07.ourcompany.com
    /_layouts/images/day.gif
    1,051
    max-age=31536000
    image/gif
    iexplore:15344
    192
    200
    HTTP
    sp07.ourcompany.com
    /_layouts/images/month.gif
    1,068
    max-age=31536000
    image/gif
    iexplore:15344
    193
    200
    HTTP
    sp07.ourcompany.com
    /_layouts/1033/styles/calendar.css?rev=BrbrIU86qTG2EHx1ZUuFBQ%3D%3D
    28,814
    max-age=31536000
    text/css
    iexplore:15344
    194
    200
    HTTP
    sp07.ourcompany.com
    /_layouts/images/week.gif
    1,057
    max-age=31536000
    image/gif
    iexplore:15344
    195
    200
    HTTP
    sp07.ourcompany.com
    /_layouts/images/weekbox.gif
    149
    max-age=31536000
    image/gif
    iexplore:15344
    196
    200
    HTTP
    sp07.ourcompany.com
    /_layouts/images/alldayDefault.gif
    157
    max-age=31536000
    image/gif
    iexplore:15344
    197
    200
    HTTP
    sp07.ourcompany.com
    /_layouts/images/calnumBttntoday.gif
    146
    max-age=31536000
    image/gif
    iexplore:15344
    198
    200
    HTTP
    sp07.ourcompany.com
    /_layouts/images/calnumBttn.gif
    95
    max-age=31536000
    image/gif
    iexplore:15344

    Hi
    when you request a  Site collection (http://domain/) or a Site (http://domain/foo/) of your Publishing Site you get redirected to the http://domain/Pages/<WelcomePage>.aspx. SharePoint 2007 uses the 302 header (location temporarily moved) for
    this purpose. Surprisingly even WSS uses the 302 header to redirect a root url to the default.aspx. In comparison ASP.NET uses an internal redirect to render the default page when the root url requested: there is no redirect in this situation.
    Check the link which can give more input
    http://blog.mastykarz.nl/sharepoint-2007-redirect-solved-using-301-instead-of-302-redirects/
    for error 401 you can check this link
    http://discussions.citrix.com/topic/97027-no-resources-401-unauthorized/d
    and check SharePoint ULS logs and event viewer for any exceptions.
    Please mark the Answer and Vote me if you think that it will help you to resolved your issue

  • Font rendering change after upgrade

    After upgrading freetype from current to the latest version I noticed that my fonts changed. The rendering was thicker, as in most other linux distributions. Not that it was bad, but I prefer the thin ones Arch had up to now.To solve it I just downgraded freetype to the previous version and everything is fine, but I wonder if the change is due to the way both versions are configured/compiled or is it an internal change in freetype?
    I ask because I'd like to keep this thinner fonts, so I want to know if I should just stop upgrading freetype forever or could I recompile the latest version to get this kind of fonts?

    Bogart wrote:
    slackhack wrote:btw, do you have TFT or CRT? i have a TFT and have tried autohinting both on and off, but it looks like total crap with it off.
    I have a CRT, since TFT screens hurt my eyes with their crispness. I prefer the smooth CRT image. I use the defult hinting (medium), though I've found that if I turn it to full hinting I get the thinner fonts too with this latest freetype, so now I can choose without the need to downgrade.
    that must be the reason, because i have a TFT and just upgraded to the latest freetype2/fontconfig, and the fonts look worse than ever. changing hintfull to hintmedium in ~/.fonts.conf has no effect whatsoever. i moved the /etc/fonts/conf.d directory, no effect, removed local.conf, no effect. nothing seems to fix them, it's really annoying.  and worst of all, the old freetype2 isn't installing because it requires an older fontconfig, and when i try to install an older fontconfig, i get some libexpat library error and everything crashes when i try to launch it.
    >>update: i just checked in gnome, and the fonts look really good there. XFCE4 must use some different method of hinting the fonts than gnome, which i thought uses the freetype2 autohinter. i changed hinting to medium in the xfce4 UI settings, and that improved things a little, even though i have hintmedium set in ~/.fonts.conf. maybe xfce4 uses the BCI hinter and bypasses the freetype hinter?

  • IPod changed after upgrade to new iTunes

    After upgrading to latest iTunes, there were some changes to my iPod that I hate. All of my songs that were numbered or had numbers in the title are not at the bottom of the songs list instead of being the first songs listed. This is also the same for my videos. And my videos that were pretty clear are now are kind of fuzzy. These were videos that were converted into mp4 format, and most of them were downloaded from YouTube. I wonder does this have anything to do w/ the face that you can download videos from YouTube onto that **** iPhone. Oh, and the other night I had to reset my iPod and a Check with "Disk Mode" or "Now in Disk Mode" popped up. I honestly don't remember putting my iPod into disk mode. It's nothing serious, but it is just really annoying. So how do I fix this, or should I just go back to the previous iTunes?

    "All of my songs that were numbered or had numbers in the title are not at the bottom of the songs list instead of being the first songs listed."
    I assume you meant to say "now" at the bottom of the list?
    Unfortunately, like it or not this how the latest version of iTunes sorts songs. Numbers now appear at the end of the listings.

  • Adjustment effects changed after upgrades

    Here is the problem statement: the effect of certain adjustments made in previous versions of Aperture changes after some Aperture upgrades. As a result, files processed in Aperture change appearance, sometimes dramatically. Does anyone have a method of managing this problem?
    I believe that the behaviour is inevitable; Aperture applies adjustments to the orginal master at "run time". This means that as underlying processing methods are changed between different versions of the software their effect on the final outcome, especially in conjunction with other adjustments, can sometimes be dramatically different. For some adjustments, like the RAW processor, Apple provided an "Upgrade" control that allows you to decide whether or not you want to risk the upgrade of that adjustment for a particular image. However, I found that many times there are more changes that cannot be manually managed.
    When I open an older RAW file with many adjustments, for a few seconds I can still see the old preview of the image. The preview looks exactly as it should. This only lasts while the engine is recalculating and applying the adjustments according to the new method. Then, Aperture rebuilds the image preview and I can clearly see the difference (usualy the result is disastrous). Having thousands of processed RAW files, pulling up one of them feels like Russian roulette: I never know whether I lost my work or not.
    Other than exporting all processed files as TIFFs BEFORE ANY Aperture upgrade I cannot think of a reasonable way to mitigate this. Aperture vaults are useless in this case, as the new version of Aperture re-applies all adjustments according to the new methods. (Please note that versions that do not have certain Aperture adjustments applied are fine; all TIFFs processed externally are unchanged, unless they have Aperture controls applied at the end of the workflow)
    Any ideas, anyone?
    Thanks in advance,
    Sorin Vacaru

    The latest is the Highlights and Shadows adjustment. I am not 100% sure when it changed (I could find out, though), but if you open with version 3.4.3 older images that used the H&S adjustment you can actually see an "Upgrade" button in the H&S adjustment panel.
    But Apple took care to provide backward compatibility, because they wanted to ensure, that you will be able to keep the old settings. Only if you willingly press that "upgrade" button your images will be reprocessed. In my library not a single image did change.
    A clear indication of the problem that I have is that when I open and old edit I can still see the old (correct) preview, until Aperture reprocesses the adjustments and the preview. There is nothing I can do from that point onwards to recover the old edit and look.
    Have you changed the Preview settings? In the latest version you can opt to use the camera previews instead of Aperture generated previews (In the Import tab of the Preferences panel). If you toggled recently from "Camera previews" to "Standard Previews", any rebuilding of previews will change the previews you are seeing.

  • Any performance gains after upgrade to Oracle 10 ?

    Hello,
    we have been using EBS 11.5.9 (with Oracle 9). Does anybody see any improvements from performance point of view after upgrade to Oracle 10r2 (but without upgrade of EBS to R12) ?
    Regards

    Somethings have been faster for us and somethings have been slower. In general it seems the same or better. When we get a chance to look at the jobs that are performing poorly I'm sure we will be able to make them perform. Make sure you find and apply all of the performance patches for the modules you have implemented, we are missing one right now and it is hurting our Configurator developers instance. I have actually set the the optimizer_features_enable back to 9i. This is something I would not recommend doing in a production environment, and we are only doing it until we can get the patch that fixes the optimizer in 10g applied.
    One surprise for us was the increase in memory required, in our test and development instances we have found a 1GB SGA is as small as you can go with 10g and still have reasonable performance. We had a number of 9i ebus databases around 300-400MB SGA that performed just fine. This limited the number of environments we could squeeze on a single server (not really a bid deal if you only have a few dev, test and training environments).
    In our production environment in order to have the same size datacache more memory was consumed. Were we could get away with a shared pool of 4GB under 9i, it grows to over 6GB in 10g. Looks like more PGA is also being used. 10g along with RUP5 and upgrading the java from 1.3 to 1.5 has increased our memory consumption. We increased memory from 64GB to 96GB and we are using a large portion of the additional 32GB. We were right at the limit with 64GB in our 9i production instance but we had never experienced the paging that we did with 10g. Now at 96GB we nolonger see any paging.

  • Firefox/Thunderbird Font Rendering changed after upgrade of xulrunner

    Hi guys,
    I experienced a problem with font rendering in the mentioned applications after upgrading xulrunner from 1.9.2.10-1 to 1.9.2.10-2 today. It seems like font smoothing/subpixel rendering does not work as before. After downgrading to 1.9.2.10-1, it worked as expected again.
    To make it more clearly, here is a comparison shot of the two different renderings. The top one is before, the bottom one after the upgrade. You can see that the font at the top is rendered more smoothly.
    And here is a difference image:
    Is anything known about changes which could cause this?
    <edit>I have to correct this. The rendering in thunderbird was obviously not changed by the xulrunner upgrade, but by a separate thunderbird upgrade from 3.1.4-1 to 3.1.4-2.
    <edit2>Additional information: I have the cleartype versions of freetype2, cairo and libxft installed, if that matters.
    Last edited by Singul (2010-09-28 16:40:18)

    Although I am empathetic to the hard (and really great) job the arch devs do, I want to voice my overall agreement with softtower on this issue. Fonts have been an off and on problem for me with arch since the beginning. I understand the devs are doing the best they can, especially when the actual problem is with the moz. devs. But fonts are the major way most users interact with the system, and I can tell you that almost nothing is more aggravating, to me at least, to suddenly "upgrade" and get presented with those really crappy fonts out of the blue. Just saying I understand the frustration, and I think he makes some good points about shipping broken packages and "blaming" it on upstream. If they're broken - don't use them! if possible. "Progress" that breaks stuff isn't really progress, imo.
    hokasch wrote:
    Don't quite get what the fuzz is all about, my fonts look as crisp as ever (on a laptop lcd), the update changed absolutely nothing for me. I never used "fixed" versions of anything, can't really remember how I set up the fonts when I did this installation though (font config advice on the wiki I guess).
    It is not the repo's maintainers job to keep packages synced up with any "fixed" packages in AUR, that is your own responsibility, and bitching about it won't help you any.
    Just pointing out that comments like these aren't really helpful, imho. So you didn't have the problem - that's good to know, but the "fuzz" is about the fact that some of us *DO* have the problem.
    It typically doesn't show up on laptops, btw, only standalone LCDs for some reason (has always been the case for me, at least, with both LCDs and laptops, where the laptop never had the problem). So again, the fuss is that a major usability function got broken from an upgrade that shouldn't get broken, and imo it raises some good, if admittedly difficult to resolve, points, philosophical points, etc. about the packaging and upgrade process,

  • HBA controller name changes after upgrading to solaris 11.2 from solaris 11.1

    we have upgraded OS on control domain from Sol 11.1 to Sol 11.2 . After upgrade controller name of HBAs and NPIV wwns  changed ( like if it is c2 prior to upgrade  , post upgrade controller name changes to  c8). Due to this we have to redo virtual disk mapping for all ldoms hosted on this machine. .  Would it be possible to preserve the HBA controller names across OS versions ? we even observed this issue while upgrading SRU of solaris 11.2 system.  We have to upgrade around 20 systems to solaris 11.2 and with this issue upgrade seems to be a tedious task. Any idea if there is any fix or workaround for it ?

    Are your disks only connected to one single HBA?
    We use MPXIO to MultiPath over Multiple HBAs. Controller never changed here for the MPXIO devices.
    Best regards,
    Marcel

Maybe you are looking for