Max number of instances per Server

Good morning,
I've installed Oracle9i in a Windows 2000 Server with 1GB of memory. After that I've created several instances of Oracle DB. With 3 instances my server works acceptably, but after adding two mor instances I've begun to have some perfomance problems and the access speed has considerably decreased, so my question is, what is the maximum number of instances that the same Oracle server can support in order to optimize the access to the DB?
Thanks in advance.
Isabel

Hi,
I am confronted with a similar problem. ie, we have got several instances on the basis of one application/one instance. I am asked to regroup all these instances to one and only one instance containing several schemas to reduce the number of servers, number of instances and on the way reduce the administrative ( dba ) tasks. As i have got some knowledge of dba, I thought of installing this instance under Oracle 10g like this : an application on its own tablespace and the backup of this tbs via rman. So when an application crashes, I can make use of TSPITR to bring the database to its initial ( stable ) status. My questions are :
1° any re-initialisation of the database will sanction all the schemas. Is there any
way to avoid this ?
2° as different tablespaces be brought to a consistency status different in time via
TSPITR even if the controlfile happens to be the same ?
3° if I opt for UNDO management AUTO, there can be only one UNDO TBS
active at a time. As this won't render me to trouble when i need to recover
only one application ? or on performance ?
Any prompt reply will be greatly appreciated.
Best regards.
Tiroumalai

Similar Messages

Trickles connection; max number of clinet per server; more services

Dear,
I found on web something about Trickles connection. Is there some further reading about Trickles connection in Java? Is that just logical concept or it is language-oriented implementation? If it is only logical concept, I suppose it is equal to TCP connection on client side and UDP connection on server side, of course if that is possibly? Can anyone tell me something more about that? Thank you.
If I am using stateful connection between server and client, what is average number of client which server can serve - you can specify some PC configuration and give number which depends on that PC? Thank you.
If somewhere are more services which should communicate, but they are on a same PC (suppose all are written in Java). What is the best way of communication between themselves?
If you can, recommend me some very good book which can give me at least one answer on my questions. It is better, if book is available as e-book.

It's not supported in Java so you're asking in the wrong place.

ISV embedding database with one instance per server limitation

Hi All,
As an ISV, it would be nice for Oracle to provide a better embedding solution than a silent install then having to script the creation of, or restore the database. Not because it's too complicated for an ISV to script, but purely for speed reasons. It's much faster to copy pre-loaded data files onto a machine than to create empty data files, load the data in, perform an analyze and any other scripts on the database. Dems just my thoughts.
However, my real issue is regarding the 'one instance per server' install limitation. As an ISV, if I choose to embed Oracle XE in my application - how can I 'guarantee' that my product can be installed on any given server. Picture this:
- We submit a proposal/ tender for a license to our software.
- Client, in conjunction with their IT dept adds internal costs, identifying servers that will run on, responsibilities, project plans (the whole works) etc.
- Client gets signoff from their management and pays for license.
- Client goes to install our software using Oracle XE just to find that another ISV has their software with Oracle XE embedded installed on the server.
- Client either has to find/ buy another server, or we lose the license. Either way, client is unhappy.
Am I missing something here? Is anyone else anticipating this problem?
Brad

Hi Florin,
Thanks for the response.. a good suggestion and our schema would be < 1GB so size would hopefully not be a problem. However, as soon as we plug into another appliation's Oracle XE implementation - we've potentially lost control of the database and it loses it's "embedded" nature. Probably not a huge problem if it's an application independant implementation of Oracle XE by an IT person, but if it's another ISVs embedded database - the dramas could start.
We effectively put our user's database within the control of another application who may not have considered the possibility of our data residing there. eg. They could do a 'drop-all' and we lose our database. Or tune the database right down and our application suffers. Or the user uninstalls that application and it decides to automatically remove Oracle XE (after all, that ISV thought it was an embedded application, right?). Do you see the potential issues?
Perhaps Oracle XE isn't as attractive for ISVs as first thought... I guess at the heart of the issue is the fact that Oracle XE isn't really an embedded database. It's just a database you can [silently] install with your application and then becomes a general db on the machine.

Max. number of WLCs per WCS

Hi,
I know this is a fairly basic question but I can't find answer in WCS Q&A's etc.
I'm looking at a deployment where there will be many small sites with only 5 or 6 APs and a 2106 WLC. Each site will be in a different mobility group.
I know WCS licensing counts APs, but is there a max. number of WLCs?
Thanks, MH

Hi Mark,
The supported number of WLC's per WCS is based on the WCS Platform used :)
Minimum server requirements
Cisco WCS High-End Server
â¢ 3000 lightweight access points, 1250 standalone access points, 750 wireless LAN controllers
â¢ Two IntelÂ® Xeon Dual Core CPU's; 3.0 GHz, 8 GB RAM, 200 GB HDD
Cisco WCS Standard Server
â¢ 2000 lightweight access points, 1000 standalone access points, 150 wireless LAN controllers
â¢ IntelÂ® Dual Core CPU; 3.2 GHz, 4 GB RAM, 80 GB HDD
Cisco WCS Low-End Server
â¢ 500 lightweight access points, 200 standalone access points, 50 wireless LAN controllers
â¢ IntelÂ® CPU; 3.06 GHz, 2 GB RAM, 30 GB HDD
CiscoWorks WLSE Models 1130-19 or 1133 running Cisco WCS
â¢ 1500 lightweight access points, 100 wireless LAN controllers
â¢ Intel Pentium 4 CPU; 3 GHz, 3 GB RAM, 38 GB HDD
http://www.cisco.com/en/US/prod/collateral/wireless/ps5755/ps6301/ps6305/product_data_sheet0900aecd802570d0.html
Hope this helps!
Rob

Max number of VMs per free ESXi hypervisor?

VMware Community,
I've just installed the free ESXi 5.5 hypervisor and I'm wondering if the same virtual machine limits apply to the standalone scenario as vCenter? Can I actually run 512 VMs? What, if any, is the max number of VMs I can run on a free version of ESXi?

As of ESXi 5.5 most limitations were removed and only an 8 vCPU limit per virtual machine remains. So unless I'm mistaken, the mentioned 512 VM per host (see http://www.vmware.com/pdf/vsphere5/r55/vsphere-55-configuration-maximums.pdf) apply to the free Hypervisor too.
André

Maximum number of targets per server ?

As I am currently struggling with configuring an agent and the monitoring template such that it can handle a strained server running a high number of instances, I was wondering what others are dealing with, and if I am not trying to accomplish more than is possible.
The server running the agent has some 135 instances running, and I am having big time delays in discovering new targets, configuring the credentials and seeing the correct status appear after agent restart or blackout ends. Also, the agent recycles itself when metrics cannot be collected on time (skipped) for multiple times, or APIs timeout
Anyone experience with agents on such loaded servers?

Awaiting further action from support, I am also investigating the possibility of installing an additional agent, but according to note 314310.1 (http://metalink.oracle.com/metalink/plsql/ml2_documents.showFrameDocument?p_database_id=NOT&p_id=314310.1) It is not possible, as the second agent will corrupt the repository because of being the same host ...
Or am I missing something ?

Clustering w/ Multiple Instances per server

For performance reasons, we want to setup multiple instances of weblogic on each server in our cluster configuration.How is this handled in a cluster configuration? Is each instance considered another server in the cluster? Does each require its on IP address, or is a seperate port sufficient for each? Also, is there a way to configure the cluster w/o using multicast - there's issues with supporing multicast on our backend network.Thanks, Rick


Sure, if JVM uses native threads and the application itself is well-behaved
          concurrency-wise (very important!) it should be able to scale up on multiple CPUs.
          Madhu <[email protected]> wrote:
          > What about CPU utilization? What if you have mutliple cpus? Is the Java VM
          > optimized for mutliprocessor environment. I can VM per processor so that I
          > can keep them all busy? Is this possible? Can I have a VM locked on to one
          > processor?
          > Also, I found issues running code (could be buggy testing) running on WL 5.1
          > on multi-cpu machine, on single cpu it was all code and in mutli-cpu, mostly
          > index out of bound / some from java.util classes like ArrayList? Is this
          > common?
          > - Madhu
          > "Dimitri Rakitine" <[email protected]> wrote in message
          > news:[email protected]...
          >> Sure you can - each instance will need it's own IP address, but why do you
          > think that running WLS
          >> instance per CPU will improve performance? The reason for running multiple
          > JVMs on the same box was
          >> memory utilization, which is not an issue with Hotspot JVM.
          >>
          >> Richard Kemp <[email protected]> wrote: > For performance reasons, we
          > want to setup multiple instances of weblogic on
          >> each server in our cluster configuration.How is this handled in a cluster
          > configuration? Is each instance considered another
          >> server in the cluster? Does each require its on IP address, or is a
          > seperate port sufficient for each? Also, is there a way
          >> to configure the cluster w/o using multicast - there's issues with
          > supporing multicast on our backend network.Thanks, Rick
          >>
          >> --
          >> Dimitri
          Dimitri

Max number of instances for stateful session beans

Hi,
I am using OC4J 9.0.3 which is embedded in JDeveloper. I would like to set the maximum number of instances for stateful session beans. I know that the session-deployment section of the orion-ejb-jar.xml file has an attribute called max-instances, but it only works for stateless session beans. (at least it says so in the documentation and it did not work with my stateful session bean)
Anyone knows how to do it?
Thanks,
Leonardo Penha

Leonardo -- Stateful session beans do not come from a pool but are assigned 1 bean to 1 user. When the EJBs life ends the bean goes away. If what you are asking about is passivation configuraiton then that is different. We currently do not support passivation based on some count of SFSBs that exists inside the container but we will provide that capability in the next release of OC4J. (Note: When passivation occurs is mostly up to the vendors and it is not required by the J2EE 1.3 specification based on performance metrics. What is required is that when a SFSB is passivated it is done correctly and we do that.)
Thanks -- Jeff

Throttling a file adapter to consume a max number of files per minute

Is there a way to design a file adapter to throttle its processing bandwidth.
A simple use case scenario is describes as follows;
A File Adapter can only consumes a max of 5 files per minute. The producer average throughput is 3 files per minute but during peak times it can send 100 files per min. The peak times occur during end of year or quarterly accounting periods. If the consumer consumes more than 5 files per min the integrity of the environment and data is compromised.
The SLA for the adapter is to :
- Each file will be processed within 2 seconds.
- Maximum File Transactions Per minute is 5
An example is as follows.
The producer sends 20 files to its staging directory within a minute. The consumer only processes 5 of these files in the first minute, sleeps then wakes up to consume the next 5 files in the second minute. This process is repeated until all files are processed.
The producer can send another batch of files when ever it likes. So in the second minute the producer can send another 70 files. The consumer will throttle the files so it only processes 5 of these files every minute.

Hi,
If you have a polling frequency set to 2 secs..then controlling it to read only five files a min is difficult.?You can have the polling frequecy changed to 12secs
or u can schedule a BPEL for every min and use synchronous read operation and loop it 5 times to read 5 files.

Max Number of threads per Process in Solaris 8

I'm running into qpplication which crashes with a segfault when the LWP for that process hits about 255, i'm trying to find the maximum number of threads per process using sysconf and its returning -1
#include <unistd.h>
int
main(void)
printf("%ld\n",sysconf(_SC_THREAD_THREADS_MAX));
return 0;
I'm using gcc 3.3.2
If anyone can tell me the maximum number of threads per process I would really appreicate it.
Also the maximum number of open files per process.
Thanks in advance
- Rodrick Brown
- DoITT

Please make sure that you build a
multithreaded program first (check with ldd).
If anyone can tell me the maximum number of threads per process I would
really appreicate it.
Also the maximum number of open files per process. It depends on the architecture (x86/SPARC), OS version,
64 or 32-bit, /etc/system, shell limitations (/usr/bin/ulimit) ...
Search for "rlim_fd_max / rlim_fd_cur" on docs.sun.com too.
HTH,
-vladimir

Max number of agents per proxy

Is there a suggested maximum number of agents per proxy, especially a proxy co-located with the enterprise controller?
Thanks,
Tim Reidy

This would depend on usage patterns and what type of hardware the proxy controller is running.
According to some whitepapers I have you're looking at :
For only hardware management - 5000 systems per enterprise controller, 1000 systems per proxy controller
For hardware, OS, update and virtualization - 3000 systems per enterprise controller, 1000 systems per proxy controller
hardware specs are:
Sparc T3's, 16-core T3 processor, 32 GB RAM for the EC, 16 GB RAM for ProxyController
Intel, Sun Fire x4270's, 2 x Quad-core Nehalem processors, 36 GB RAM for EC, 8 GB RAM for ProxyController

Number of applications per server?

Just curious... how many applications do you run per iAS? At the moment we have only one application per server, which seems like a waste of resources to me (considering the average CPU load is less than 1 %!).
And a 'side question' - do you create more OC4J:s for every added application or do you let one take care of all applications?

Hi Magnus,
it's an interesting question and i hope to see some feedback from users around the world.
Im my server (hp proliant dl580, windows 2008 , Oracle App Server 9.04, 4GB Ram) i have deployed 8 applications (planning to deploy one more till end of month)
i've created 3 containers (OC4J) , one OC4J for each branch of business, ie, i'm working for an insurance company , so the distribuition is the following
OC4J A - Application for "Real" Insurance (car, houses,etc)
OC4J B - application for "Non Real" insurance (life insurance,etc)
OC4J C - other general application
Fortunatelly ,the most resource intensive application (memory, cpu , database connections) are in OC4J A and OC4J B
So, as a best pratice i advice to split applications across containers by availability and resource consuption
Best Regards
Rui Madaleno

CPU_COUNT when more than 1 instance per server

Hi,
I wonder if there is any drawback when we got few instances on the same server from cpu_count point of view.
This parameter is derived by all instances like they are alone so all control structures are allocated independetly but there is more
resource needed when many instances in place.
Could this be potencial bottleneck ?
Regards.
Greg

mseberg wrote:
The last time I checked the guideline the maximum CPU busy rate should be 90%
Here's more info:
http://www.dba-oracle.com/t_high_cpu.htm
The advice on that page is not very good - the most important point is the quote from IBM, which highlights the fact that there is a difference between single-user and multi-user systems:
<blockquote>
+"This holds true in the case of a single-user system with no need to share the CPU. Generally, if us + sy time is below 90 percent, a single-user system is not considered CPU constrained. However, if us + sy time on a multiuser system exceeds 80 percent, the processes may spend time waiting in the run queue. Response time and throughput might suffer"+
</blockquote>
In fact, it's a bit silly to give a single figure for the point where you should be concerned about CPU utilisation. The best reference for discussion is Cary Millsap's book "Optimizing Oracle Performance" which discusses the impact of concurrency on queueing. Instability in performance varies with the number of CPU combined with the percentage utilisation, and the break point sets in around the following levels:
<blockquote>
4 CPUs: 67%
8 CPUs: 75%
16 CPUs: 87%
</blockquote>
Your observation, thought that the OP should not mess with the cpu_count parameter is sound. Perhaps the only parameters that he should worry about are the parallal_max_servers and processes parameters to ensure that no one instance is allowed to run up more processes than the machine can handle.
(I find the best sources of information that bridge the gap between Oracle and the hardware are Kevin Closson's blog, James Morle's website - and a site from someone at Sun whose name I can't remember at the moment.)
Regards
Jonathan Lewis
http://jonathanlewis.wordpress.com
http://www.jlcomp.demon.co.uk
To post code, statspack/AWR report, execution plans or trace files, start and end the section with the tag {noformat}{noformat} (lowercase, curly brackets, no spaces) so that the text appears in fixed format.
There is a +"Preview"+ tab at the top of the text entry panel. Use this to check what your message will look like before you post the message. If it looks a complete mess you're unlikely to get a response. (Click on the +"Plain text"+ tab if you want to edit the text to tidy it up.)
+"Science is more than a body of knowledge; it is a way of thinking"+
+Carl Sagan+

Multiple instances per server

Back in January, I talked to Oracle about multiple instances on NT. They said I could do it but not recommended for production.
Finally received the 11.5.3 cd kit so getting ready to do the upgrade on development. In looking at the "upgrading" chapter of the "install" guide, the panel displays only show information for one instance!!! How do I enter information for other instances including a new demo database. Back in January, I was told I could share the appl_tops.

Back in January, I talked to Oracle about multiple instances on NT. They said I could do it but not recommended for production.
Finally received the 11.5.3 cd kit so getting ready to do the upgrade on development. In looking at the "upgrading" chapter of the "install" guide, the panel displays only show information for one instance!!! How do I enter information for other instances including a new demo database. Back in January, I was told I could share the appl_tops.

Max number of devices per app

I work for a school and we have 200 ipads. Can I purchase one app and place it on all 200. I am assuming not, but are there restrictions?

Repost your question in the iPad in the Enterprise (Schools/Business) forum https://discussions.apple.com/community/ipad/ipad_in_the_enterprise Maybe someone from another school system will answer. Here's one link, but I'm not sure it applies to schools. http://www.idownloadblog.com/2011/08/10/apple-id-limited-10-devices/
 Cheers, Tom

Max number of instances per Server

Similar Messages

Maybe you are looking for