Shared memory on the Sparc Server running Oracle 10 G

Newbie: I can overscribe memory on a server that hosts multiple instances of Oracle? It seems normal to oversubscribe memory on multiple instances if Oracle assigns the maximum available memory on a server at the time of the request.
For example:

How Oracle manages memory if I have 16 GB of physical memory, run 4 instances of Oracle on my server and specify the max_sga than 8 GB in my/etc/project file? This configuration seems to oversubscribe memory twice, 32 GB memory requested on a server that has only 16 GB of physical memory.

Oracle manages memory differently if I have 16 GB of physical memory, run 4 instances of Oracle on my server and specify the max_sga like 16 Gb in my/etc/project file? This configuration seems to oversubscribe memory four times, 64 GB memory requested on a server that has only 16 GB of physical memory.


Thank you for your help.

user12188192 wrote:
Oracle manages memory differently if I have 16 GB of physical memory, run 4 instances of Oracle on my server and specify the max_sga like 16 Gb in my/etc/project file? This configuration seems to oversubscribe memory four times, 64 GB memory requested on a server that has only 16 GB of physical memory.

With Solaris and DISM, you can set the MAX_SGA_SIZE and the TARGET_SGA_SIZE of different values. The TARGET_SGA_SIZE for all MORE the PGA_AGGREGATE_TARGET instances for all instances should be around 80-90% of the total RAM. You can then change the targets to let different instances have 'more' when they need while reducing others.

In this way, you can manage the memory and avoid swapping, because when it comes to performance (and managers yell at you for low performance) swap is not a good thing.

Note that this is possible on the machines and operating systems that support the DISM and enabled/configured. Which is discussed in the manuals of Solaris.

But in the ideal, a machine will have only one instance of database and several instance of database requirements will be consolidated as much as possible. (Although he explain literally hundreds sof reasons do not consolidate - several policies rather than technical.)

Tags: Database

Similar Questions

  • Move the database to the server running Oracle 11 G to the server running Oracle 12 c

    I'm trying to find the best way to migrate a database to a server that is running Oracle 11 G on a server that is running Oracle 12 c.

    I tried using the RMAN duplicate command of database but have run in many too many problems when you try to configure both servers before running the command duplicate.  If someone could provide some guidelines on configuring two servers of the RMAN Duplicate command clearly then that would be great.  The other thing, I've tried makes a cold backup of all files for the database (data files, control files, etc.).  Copy them to the new server, and then re-create the control file to include the correct location and names for the data files and the redologs.  After you re-create the database control file does not start.  I suspect that it is because of differences in version.

    If you can provide me with claire (ie. migration for Dummies) instructions to get one of these methods of work, it would be greatly appreciated.  I am also open to any other method you can think of to achieve what I'm trying to accomplish.

    Thank you

    Paul Noyes

    Paul,

    Please, go to the PDF below he explains the process of upgrading.

    You can create all the tablepsaces and then use utility Import/export where the directory paths are different

    http://www.Oracle.com/technetwork/database/upgrade/upgrading-Oracle-database-WP-12C-1896123.PDF

    Thank you

    SK

  • Running an ADF base application fails on the remote server run. Running locally on Jdeveloper 12.1.3 works very well

    I'm deploying an ear with two war files file. One of the web applications in this EAR file acts as a basic authentication web application. I am able to do basic http very well, but when the welcome.jsf try to load, this error occurs:

    ADF_FACES-30200: for more information, see the error log of the server for an entry beginning with: The UIViewRoot is null. Fatal error during PhaseId: RESTORE_VIEW 1.

    Additional detail

    java.lang.NoClassDefFoundError: oracle, adf, model, RegionBinding

    at oracle.adfinternal.controller.faces.lifecycle.ADFPhaseListener.getLifecycleContext(ADFPhaseListener.java:175)

    to oracle.adfinternal.controller.faces.lifecycle.ADFPhaseListener$ PhaseInvokerImpl.dispatchAfterPagePhaseEvent (ADFPhaseListener.java:302)

    at oracle.adfinternal.controller.faces.lifecycle.ADFPhaseListener.afterPhase(ADFPhaseListener.java:88)

    at oracle.adfinternal.controller.faces.lifecycle.ADFLifecyclePhaseListener.afterPhase(ADFLifecyclePhaseListener.java:53)

    at oracle.adfinternal.view.faces.lifecycle.LifecycleImpl._executePhase(LifecycleImpl.java:489)

    at oracle.adfinternal.view.faces.lifecycle.LifecycleImpl.execute(LifecycleImpl.java:225)

    at javax.faces.webapp.FacesServlet.service(FacesServlet.java:593)

    to weblogic.servlet.internal.StubSecurityHelper$ ServletServiceAction.run (StubSecurityHelper.java:280)

    to weblogic.servlet.internal.StubSecurityHelper$ ServletServiceAction.run (StubSecurityHelper.java:254)

    at weblogic.servlet.internal.StubSecurityHelper.invokeServlet(StubSecurityHelper.java:136)

    at weblogic.servlet.internal.ServletStubImpl.execute(ServletStubImpl.java:346)

    at weblogic.servlet.internal.TailFilter.doFilter(TailFilter.java:25)

    at weblogic.servlet.internal.FilterChainImpl.doFilter(FilterChainImpl.java:79)

    at oracle.adf.share.http.ServletADFFilter.doFilter(ServletADFFilter.java:77)

    at weblogic.servlet.internal.FilterChainImpl.doFilter(FilterChainImpl.java:79)

    at oracle.adfinternal.view.faces.webapp.rich.RegistrationFilter.doFilter(RegistrationFilter.java:105)

    to org.apache.myfaces.trinidadinternal.webapp.TrinidadFilterImpl$ FilterListChain.doFilter (TrinidadFilterImpl.java:502)

    at oracle.adfinternal.view.faces.activedata.AdsFilter.doFilter(AdsFilter.java:60)

    to org.apache.myfaces.trinidadinternal.webapp.TrinidadFilterImpl$ FilterListChain.doFilter (TrinidadFilterImpl.java:502)

    at org.apache.myfaces.trinidadinternal.webapp.TrinidadFilterImpl._doFilterImpl(TrinidadFilterImpl.java:327)

    at org.apache.myfaces.trinidadinternal.webapp.TrinidadFilterImpl.doFilter(TrinidadFilterImpl.java:229)

    at org.apache.myfaces.trinidad.webapp.TrinidadFilter.doFilter(TrinidadFilter.java:92)

    at weblogic.servlet.internal.FilterChainImpl.doFilter(FilterChainImpl.java:79)

    to oracle.security.jps.ee.http.JpsAbsFilter$ 1.run(JpsAbsFilter.java:137)

    at java.security.AccessController.doPrivileged (Native Method)

    at oracle.security.jps.util.JpsSubject.doAsPrivileged(JpsSubject.java:315)

    at oracle.security.jps.ee.util.JpsPlatformUtil.runJaasMode(JpsPlatformUtil.java:460)

    at oracle.security.jps.ee.http.JpsAbsFilter.runJaasMode(JpsAbsFilter.java:120)

    at oracle.security.jps.ee.http.JpsAbsFilter.doFilter(JpsAbsFilter.java:217)

    at oracle.security.jps.ee.http.JpsFilter.doFilter(JpsFilter.java:81)

    at weblogic.servlet.internal.FilterChainImpl.doFilter(FilterChainImpl.java:79)

    at oracle.security.wls.filter.SSOSessionSynchronizationFilter.doFilter(SSOSessionSynchronizationFilter.java:194)

    at weblogic.servlet.internal.FilterChainImpl.doFilter(FilterChainImpl.java:79)

    at oracle.dms.servlet.DMSServletFilter.doFilter(DMSServletFilter.java:220)

    at weblogic.servlet.internal.FilterChainImpl.doFilter(FilterChainImpl.java:79)

    at weblogic.servlet.internal.RequestEventsFilter.doFilter(RequestEventsFilter.java:27)

    at weblogic.servlet.internal.FilterChainImpl.doFilter(FilterChainImpl.java:79)

    to weblogic.servlet.internal.WebAppServletContext$ ServletInvocationAction.wrapRun (WebAppServletContext.java:3436)

    to weblogic.servlet.internal.WebAppServletContext$ ServletInvocationAction.run (WebAppServletContext.java:3402)

    at weblogic.security.acl.internal.AuthenticatedSubject.doAs(AuthenticatedSubject.java:321)

    at weblogic.security.service.SecurityManager.runAs(SecurityManager.java:120)

    at weblogic.servlet.provider.WlsSubjectHandle.run(WlsSubjectHandle.java:57)

    at weblogic.servlet.internal.WebAppServletContext.doSecuredExecute(WebAppServletContext.java:2285)

    at weblogic.servlet.internal.WebAppServletContext.securedExecute(WebAppServletContext.java:2201)

    at weblogic.servlet.internal.WebAppServletContext.execute(WebAppServletContext.java:2179)

    at weblogic.servlet.internal.ServletRequestImpl.run(ServletRequestImpl.java:1572)

    to weblogic.servlet.provider.ContainerSupportProviderImpl$ WlsRequestExecutor.run (ContainerSupportProviderImpl.java:255)

    at weblogic.work.ExecuteThread.execute(ExecuteThread.java:311)

    at weblogic.work.ExecuteThread.run(ExecuteThread.java:263)

    Caused by: java.lang.ClassNotFoundException: oracle.adf.model.RegionBinding

    at weblogic.utils.classloaders.GenericClassLoader.findLocalClass(GenericClassLoader.java:357)

    at weblogic.utils.classloaders.GenericClassLoader.findClass(GenericClassLoader.java:318)

    at weblogic.utils.classloaders.ChangeAwareClassLoader.findClass(ChangeAwareClassLoader.java:80)

    at java.lang.ClassLoader.loadClass(ClassLoader.java:425)

    at java.lang.ClassLoader.loadClass(ClassLoader.java:358)

    at weblogic.utils.classloaders.GenericClassLoader.loadClass(GenericClassLoader.java:186)

    at weblogic.utils.classloaders.ChangeAwareClassLoader.loadClass(ChangeAwareClassLoader.java:50)

    ... more than 49

    This problem does not occur on my installation local jdeveloper (12.1.3)

    Deploy the application to a 12 c Enterprise development environment manager, and there is only one target server.

    Have you checked that the target server has the right adf 12.1.3 installed on?

    The error normally means that the runtime is missing.

    Timo

  • Change the shared memory in the Bios for the graphics - Satellite C660 217

    There is no setting in the BIOS that I can find.

    BIOS is up to date

    Hello

    It is not possible.

    Shared memory is assigned automatically and this is controlled by the graphics driver.
    But the shared memory is only used if the applications (IE game) would require more graphics RAM.

  • Remote access client 2.0.1 for Mac OS 10.4.11 connection to the remote server running Windows Server 2008 R2

    I have a mac with OS 10.4.11 and I try to access a remote server running Windows Server 2008 R2.  I have installed, uninstall and reinstall the Client of remote access for Mac 2.0.1 but I can not connect.  My network administrator tells me that Mac OS 10.4.11 is so old and useless that I am better just throw my computer and buy a new one in order to solve the problem.  This is absurd.  Can you help me?

    Hi Benjamin Spicer,.

    Your question is more complex than what is generally answered in the Microsoft Answers forums. It is better suited for Windows Server on TechNet. Please post your question in the Forums TechNet Remote Desktop Services (Terminal Server Services).

  • Need a SMTP service for the web server running Windows 7 Professional SP1

    The remote server administration tools package seems to work only for the management of Windows 2008. I have a web server running on Windows 7 and want to integrate the messaging service. IIS 7.5/7.0 does not include this feature. Now that are remote server administration tools installed, I tried to use the feature of IIS 6.0, however, a message is displayed to indicate that I don't have the SMTP service is installed. I checked the 'Server' SMTP under functionality after installing IIS 6.0 from the admin of remote server Pack. This does not install the feature. Could someone give information on how I can install email/mail service or function to use with my domain name on Windows 7 Professional SP1? Thank you.

    Microsoft recently released an update for the Remote Server Administration tools work with domain controllers running Windows Server 2008 R2 SP1.

    You can download it at the following link:
    http://www.Microsoft.com/downloads/en/details.aspx?FamilyId=7d2f6ad7-656b-4313-A005-4e344e43997d

    For further assistance, please post in the following forum:
    http://social.technet.Microsoft.com/forums/en/category/w7itpro/

  • Hot cloning - let the physical server running then?

    I have a question for the Vmware Converter. If I understand correctly, you can use hot Clone to clone a physical server while it is running. Of course, that seems very well so you don't get no downtime during conversion.

    But, if the reason for this conversion is to migrate from physical to virtual, what happens if the physical server continues to operate after the conversion? Even just a minute can be enough to have changed the original system, perhaps a database or other stuff, that will make the new virtual instance "incorrect."

    So how does this normally? If you want to virtualize a server that has a system that interacts with users either or other servers, how will stop instantly change?

    OK, even if you are in your downtime window it would be preferable to perform the validation of the server application.

    What we normally do, is check the event logs and do well for sure there no new errors that are produce after the p2v (physical 2 virtual), especially application or system related event.  We will also perform a few reboots of the (virtual) system and sometimes we also had to remove some old drivers.  We have sometimes came across problems with NETWORK cards keeping the old ipconfig when migrating.

    When supply us the new VM give us it a whole new value of intellectual property and test it with the other physical system down.  After correct validation go us ahead and leave the virtual running.  We usually have to master the physical machine in the datacenter for about two weeks and then begin the process of decommissioning.

    One thing most definitely, you want to do is make sure to install vmware tools on the virtual server.  I know it's probably obvious, but many people forget.

    Hope this helps,

    Chad King

  • Column headings missing when the SQL report runs Oracle Conc ask

    Why the column headers for a sql report do not disappear when I try to run Oracle Manager of competing application work? If I copy and paste work in SQL * more and run it, the topics show on the report. Is there a setting I can update to allow these column headers to display?
    Example of what I get (No Headings)
    
    TOM SMITH          12345
    JANE DOE           34355
    
    Example of what I expect (with Headings)
    
    EMPLOYEE           EMPLOYEE NO
    TOM SMITH          12345
    JANE DOE           34355
    Thank you

    More likely:

    SET THE POSITION

    into force or / and

    SET PAGESIZE 0

    is in effect.

    SY.

  • Failure of advanced/shared navigation of the actions to run when assigned by leaving. Why?

    After numerous tests, it seems that the bottom line is as described in topic - too far go that responsive projects. The question is not present for the empty projects I can say. I am running 8.0.1.242 on Windows 8 Captivate.

    Maybe I can work around this problem by writing in a JS event listener, but I see no other way to have trigger actions conditional navigation to specific times instead of via the interaction of the user - that is a user clicks on one of the 2 icons on slide 1 to view more details, taking to the slide where this detail is covered , after which they returned to slide 1, UNLESS they saw all the slides in detail, in which case they are sent to the end/next section slide.

    Is this a known bug? Is it still a bug? You can reproduce it? See nature projects reactive with advanced navigation actions attributed to work out very well for you? Any information would be appreciated, I would go to the bottom of this. I'd be happy to post more details if this case works for you - it looks like a bug here if so I prefer not to bother in the first place.

    Thank you!

    Hi Lieve,

    Thank you for your response. I saw your article mention - you have an great Google visibility when it comes to Captivate. It's a shame that you are one of the only good resources that - and Rod of course (I know you're out there too). It seems rather odd to me that always Captivate gets out with out-of-the-box features that do not work, and support is pretty limited. I can usually find a way around them, but it always adds a lot of time to my projects of debugging to find that the problem is of Captivate and not mine... not ideal. Makes me wonder if Adobe requires releases annual no matter where is their Cp dev team.

    In this case, the same on the output function that failed in my responsive project worked when assigned in the same way in a project of standard and white - at least in this one case alone is not a general question "exit does not work as I want to." I'm not trying to run everything must happen ON the slide came out, I just need to run - at some point. I determined that all sorts of actions work in my project for leaving: I can increment variable, show and hide objects, etc., etc. - the only thing falling down (and it will fail 100% of the time for me in reactive projects) use navigation in the context of shared or advanced actions, if go to slide and other slides at base of navigation , or GotoFrameAndResume or another framework based "micronavigation" (thank you for giving me the initial end on this point, moreover - it has served me well).

    Unfortunately, I could not add the action for my enter on to the slide later, as in my case, I need to my function leaving kicker users back to the beginning or to the slide at the end of the project, so if to sign they would ever get see the content of the slide. I end up doing something similar: adding brief slides 'redirect' between my content slides, that have the same feature set to enter. Looks like a stupid work-around for me, but my JS solutions were not well behave and I got tired of debugging.

    If you want to, I would be grateful if you could check if it seems like a bug you as well and maybe even if it persists in Captivate 9 (I recall that have this you). I still get nagging that I'm something I just straight up found no clearing of doubts, although this one seems, as it has nothing to do with me. Thanks again.

  • The query will run Oracle 10 g?

    Hello

    This is the structure of the table emp and dept. There is a request on this basis, please let me know if this request is going to work?

    EMP table
    =========
    EMPID NUMBER PRIMARY KEY,
    NAME
    ADDRESS

    DEPT table
    ==========
    DEPTNO NUMBER,
    DEPTNAME

    Select * from emp where empid in (select empid from dept where dept_no = 10);

    Nordik wrote:
    Hello

    This is the structure of the table emp and dept. There is a request on this basis, please let me know if this request is going to work?

    EMP table
    =========
    EMPID NUMBER PRIMARY KEY,
    NAME
    ADDRESS

    DEPT table
    ==========
    DEPTNO NUMBER,
    DEPTNAME

    Select * from emp where empid in (select empid from dept where dept_no = 10);

    Yes, it will run.

    Oracle must first check if the EMPID column exists in the DEPT table.
    If it's not, then it checks for columns in the outer query.

    And in this case, because you do not Alias of Table. An Alias has been used, the column would have been checked only in the table with an alias.

    In addition, you will lose everything except to check the query on a database, would you?

    Pretty easy to verify this, right?

    create table test_table (col1 number, col2 varchar2(5));
    
    create table test_table2 (col1 number);
    
    select *
      from test_table t1
     where col2 in (select col2 from test_table2 t2 where t1.col1 = t2.col1);
    
    no rows returned
    
  • How to change the database server in oracle Application 11.5.10.2

    Hello

    I recently migradated my database from AIX to linux, linux and sid database port number is different name.

    How to delete and re-register the host name, the database, the port number?

    Please advice.

    Thank you
    Prince

    Prince.

    That's right a 338003.1

    It's the good doc.

    You can also check (steps to own non-existent nodes or IP addresses FND_NODES [260887.1 ID]).

    Thank you
    Hussein

  • ASIC can run on the same server/db like EBS (R12.1)

    Hello

    We have an EBS (R12.1) environment and intend to use the CPSA. I've seen environments where the CAWC runs in its own environment/database, etc. and connects to an instance of the EBS. Which is the only type of available configuration or can CAWC run fully within an environment of EBS/instance on the same server? If so, what are the consequences of such a configuration, for example on the performance.

    Thank you

    NAT.

    Hi Nat,

    ASIC can be configured in two ways:

    1 centralized - where the EBS and VCP reside on the same application server.

    2 decentralized - where the EBS and CRS are on two different application servers.

    Basically the motor nucleus of CPSA of planning does all the planning in the memory of the application server (and not in the database). This planning process is highly require a lot of resources (memory) and requires a huge memory (according to the data, the type of plan and many other factors). If the application servers are not the right size, this led to the poor performance of the application. And if both the planning and EBS reside on the same server, there are possibilities that you slow the global treatment through the functions of the company. That's why decentralized architecture is preferred, where the planning process running on another remote computer of your EBS and it didn't there in not affecting transactional modules.

    Beyond that, you have the ease of maintenance in terms of correction, update, addition of multiple of VCP modules, if you're on a decentralised architecture.

    Centralized architecture is also not rare. Many companies run their planning and EBS on a single machine.

    I hope that the above gives you a point of view, using which you can choose what suits your business best!

    Kind regards

    Mohan Balaji

  • Oracle not available, shared memory realm does not exist

    Hello

    I have problem during the installation of XE on Windows 2003 R2 SP 2 Enterprise edition. I have uninstall oracle xe and still the same result. Finally, I uninstall the windows and start all over again, surprisingly the always the same result. FYI, we have normally no problem during installation because it is part of our system, we sell to our customers who require Oracle xe. It is the first time we have seen this issue.

    ORA-01034: Oracle not available
    ORA-27101: shared memory realm does not exist.

    Oracle XE and Auditor services is started in windows services.

    If we try to start the message occurred.

    SQL > connect sysdba virtue

    Connect to an instance is idle.
    SQL > startup
    ORA-44412: setting memory XE edition invalid or unspecified
    SQL >

    I can send you the journal oracle_xe, ora init and OpsBuildAll.log to create the database.

    When compared with the installation of work, we have seen that the size of the oraclexe folder is less, it's just 233 MB while the work is beyond 1 GB. After investigation, discovered that the .dbf files is not available in the oradata folder, I tried to copy the files from one operation but the problem remains the same, no doubt there are a lot of missing more files.


    Appreciate your help on this.

    Thank you

    Agus

    It's a shame I don't remember that one. The error message (and solution) was a similar here: {: identifier of the thread = 2322695}, with a real solution from here: {message identifier: = 10039007}
    (Just to keep an eye on this kind of mistake and to link the case...)

    I'm glad to hear that you solved the issue now, but I'm still a bit puzzled how this could happen. Possibly management of memory to the point 10.2 XE does not get what he expected on systems with Hyper-Threading or several used CPU sockets, maybe is the fact, it cannot apply/monitor limitations as expected. Still the error message is a bit misleading.
    It would be interesting to know if 11.2 XE gets along on this platform. Maybe you have time to give it a try someday? I would recommend to use 11.2 XE for a new installation anyway... ;)
    It is likely to work because no one reported an error like that with 11.2 XE and there are a few positions, mainly on licensing issues, which indicate 11.2 would also go on HT and dual socket machines, but it is possible that they still had something different on their machines.

    -Udo

    P.S.:
    @Fredrik: If our messages overlap on this reference... ;)
    @Agus: If you think that your problem has been resolved, please mark this thread as answered and all useful/correct answer accordingly, for other users looking for a similar question will be able to easily spot the positions in question.
    Thank you!

    Published by: Udo on 23.03.2012 09:15

  • Question about shared memory

    Hello!

    I have a virtual machine that is configured with 1 GB of RAM and installed with Windows Server 2003. Shortly after starting the virtual machine, the info of the allocation of resources on the memory usage indicates the following:

    While the Task Manager Windows VM shows the following:

    While the use of the memory of the VM on the host is only 264 MB, why the use of the memory of comments displays 787 MB shared memory? The virtual machine is the only VM running on this host. If I understand correctly, "Shared memory" would theoretically show the amount of memory shared with other virtual machines on this host. For example, let's say I had two W2K3 VMs on that host with 1 GB of RAM each, the counter shared memory would report to the amount of memory that is the same for both these virtual machines and so physically shared. However, since there is no other virtual computer running on this host I don't really understand why he reports 787 MB!  It seems to me that the value of the shared memory is unused memory in fact (since the task inside the guest Manager reports only about 150 MB use and the virtual machine does not seem to consume the 264 MB mentioned on the host).

    Any ideas what I'm missing? Thank you!

    As you say

    • The virtual computer is configured with 1024 MB of RAM

    • The private memory usage is 237 MB

    • Shared memory is 787 MB

    All Windows VMs zero start pages is that they fill all memory with zeros. So VMKernel identifies all pages from scratch can be shared to one. In your scenario THAT VM wrote zeors 1024Mo and out of which 787 MB is shared and still 237 is used which further can be shared. 787 MB shared memory means that it the memory saved with page sharing and memory amount can be used by other processes.

    I hope this helps.

    Thank you.

  • TimesTen shared memory internal process commication

    Hi all


    My app connects to Tiemsten store data in client/server Mode using the TCP/IP protocol.
    our 32-bit Applications running on different server where the Tiemsten server running.

    I have the following questions on this

    1 can we have shared Timesten memory internal process commication to client/server architecture?

    2. What are the best practices to impelment SHMIPC in client/server mode?

    Please help me

    Thank you
    -AK

    Sharing memory Inter Process Communication (SHMIPC) is only possible when the client and TimesTen server are running on the same server. When the client and server are running on different machines that only TCP/IP is possible.

    Kind regards

    Chris

Maybe you are looking for