Managed vs Unmanaged ODP benchmarks

Has Oracle published (or does anyone have) any performance benchmarks for the managed vs. unmanaged ODP.net drivers?
Is there any performance benefit/penalty for moving to the unmanaged driver?

The answer is YMMV. Some managed ODP.NET apps will be faster and some will be slower. It depends on what your ODP.NET code does.
If ODP.NET were a car, it would be easier to provide a definitive performance measure. You drive the car in a straight line, which is generally how everyone operates a car.
ODP.NET is an API. There are many ways to use the API: how to retrieve data, data types, how much tuning is done, data characteristics, inserts/updates/selects, distributed transactions, etc. Some operations in certain code paths are faster. Some are slower. How fast your app is depends on which code paths it uses and how frequently. There is no benchmark that can provide a representative idea of what you will ultimately see in your own app unless your app is almost exactly like the benchmark, but I've never found that to be the case.

Similar Messages

What is Managed and unManaged reports

Hi ,
Can some one tell me the difference between managed and unmanaged reports that is mentioned in the RAS developer library.
thanks,
Prem

If report is Managed, you have to first publish the same report in Enterprise first and then only you can use that report inside your application.So managed report implies that Enterprise is aware of report.
If report is Unmanaged ,you can access the report by specifying its path inside your code.Unmanaged reports do not require authentication or other logon procedures to access them. Components that support unmanaged reports include the Java Reporting Component and unmanaged Report Application Servers

Summary Management and TPC-H Benchmark

I would like to simulate the smallest version of TPC-H data
warehouse with 1G data on a single machine with windows 2000
professional system platform , 40G hard drive, 256M RAM and
Pentium. III 866 processor. Currently I don't care the
performance. I just want to make it run.
I also really want to use the advisor of the Summary Management
of Oracle which is introduced in Chapter 8 of Oracle8i Data
Warehousing Guide Release 2(8.1.6).
(Actually, I have no idea about the Summary Management,
especially how to run the advisor utility which can recommend
which materialized views to create, retain and drop.)
First:
Which Oracle products do I need to download and install? I have
got Oracle8i Enterprise Edition 8.1.7 already. What others do I
need? How many servers or clients should I install ?
Second:
Is the Summary Management contained in the Server -- Typical
Installation option of Oracle8i EE(8.1.7)? How to run the advisor
Third:
Should I choose making a copy from CD to build a database ? If
not, how to configure the table space for running TPC-H which
actually is a decision support benchmark with at least 1G data..
Forth:
The 1G raw data are text files with | as the delimiter. Which
program can I use to dump the data into the database ?
I am sorry for too many questions since I am a newbie of Oracle.
Please give me any comment,suggestion and advice.
Many many thanks in advance.

This forum is for general suggestions and feedback about the OTN
site. For technical question about an Oracle product, you can
select the appropriate discussion forum in our 'Discussions'
section at: http://forums.oracle.com/forums/homepage.jsp
Regards,
OTN

Enterprise Manager on unmanaged hosts?

This is not clearly mentioned in the documentation... Does any additional software (beyond 11r2 grid infrastructure and 11r2 database software) need to be installed on a Red Hat linux el5 host to allow EM to discover it?
Basically, I wish to connect my current Enterprise Manager installation (which monitors numerous other linux systems and oracle databases) and connect a new host--as mentioned above--to it. I have tried the auto discovery method in EM already with no luck. As I am a new DBA this may be obvious.
Thank you

Sorry, I do not have quick links to it.
There are several variables that need to be set properly, but this is all within the OMS Grid installation and configuration documentation.
The easiest thing to do is just manually install the OMS agent on the server.
It is straight forward and the only thing you need is the sysman password of the OMS.
Be sure dbsnmp and sysman accounts are not locked.

Managed Dirver InputOutput Param Issue

Hi!
How do you think, can we treat behavior difference between managed and unmanaged as an issue?
Fake procedure:
PROCEDURE in_out_test( p_some_id number)
is
begin
null;
end;
Sample code:
using System;
using System.Data;
//using Oracle.DataAccess.Client;
using Oracle.ManagedDataAccess.Client;
namespace ManagedODPInOutParamTest
class Program
static void Main(string[] args)
   Console.WriteLine("go");
   try
    using (var conn = new OracleConnection())
     conn.ConnectionString = "Data Source=vos;User ID=tf2_cust;Password=bla-bla-bla;";
     conn.Open();
     using (var cmd = new OracleCommand("TEST.in_out_test", conn))
      cmd.CommandType = CommandType.StoredProcedure;
      var param = cmd.Parameters.Add("p_some_id", OracleDbType.Decimal, ParameterDirection.InputOutput);
      param.Value = 123456;
      Console.WriteLine("param val befor execute: " + param.Value);
      cmd.ExecuteNonQuery();
      Console.WriteLine("param val after execute: " + param.Value);
     conn.Close();
   catch (Exception ex)
    Console.WriteLine(ex.ToString());
   Console.WriteLine("press");
   Console.ReadLine();
Using Oracle.DataAccess we got:
go
param val befor execute: 123456
param val after execute: 123456
press
Using Oracle.ManagedDataAccess we got:
go
param val befor execute: 123456
param val after execute: null
press
I know that procedure is declared without IN OUT keyword on parameter, but the behavior of managed driver differs from the unmanaged one and this makes them not fully compatible.
Will this be fixed? Or in order to move to managed driver we have to rewrite code, that works fine with unmanaged?
PS: Server version: 11.2.0.3.0, Client version: 12.1.0.1.0, OS: Win7 x64.

And what about this, Alex:
        var cmd = conn.CreateCommand();
        cmd.Parameters.Add("ln", OracleDbType.Varchar2, 32).Value = "aaaaaa";
        cmd.Parameters.Add("dostep", OracleDbType.Byte).Direction = System.Data.ParameterDirection.Output;
        cmd.CommandText = "update uzytkownik set lastlogintime=sysdate,acccnt=acccnt+1 where lower(login_nds)=:ln returning dostep into :dostep";
In użytkownik table there is no record with login_nds == "aaaaaa"
ExecuteNonQuery() crashes with stack:
[ArgumentOutOfRangeException: Index was out of range. Must be non-negative and less than the size of the collection. Parameter name: index]
   System.ThrowHelper.ThrowArgumentOutOfRangeException() +72
   OracleInternal.TTC.Accessors.TTCNumberAccessor.GetByteRepresentation(Int32 currentRow) +28
   OracleInternal.ServiceObjects.OracleParameterImpl.ExtractIntFromAccessor(Accessor accessor, PrmEnumType enumType, OracleDbType oraDbType, Int32 currentRow) +313
   OracleInternal.ServiceObjects.OracleParameterImpl.GetIntFromBytes(Accessor accessor, PrmEnumType enumType, OracleDbType oraDbType) +279
   Oracle.ManagedDataAccess.Client.OracleParameter.PostBind_Int32(Accessor bindAccessor) +259
   OracleInternal.ServiceObjects.OracleCommandImpl.ExtractAccessorValuesIntoParam(OracleParameterCollection paramColl, OracleConnection connection, String commandText, Int64 longFetchSize, Int64 lobFetchSize, Int64[] scnFromExecution) +756
   Oracle.ManagedDataAccess.Client.OracleCommand.ExecuteNonQuery() +504
There was no record selected by a query, so returning clause couldn't return a thing
ODP.NET crashed in this scenario - this is a bug in my opinion
I'm using:
package id="odp.net.managed" version="121.1.1" targetFramework="net40" />

HT1320 hello there, i have an ipod classic (silver)120GB i have managed to download all my albums onto it.but then i have done something ,that has wiped all songs of it and now i can't redownload them .also my ipod keeps showing the apple logo and seems t

hello there, i don't have windows operating system on my computer, i have a system called ubuntu,so there for i can't download itunes. i have a music app,called banshee on my computer,and have been using it successfully.i managed to put my music from my computer to my ipod but al of a sudden it's not working and iv'e lost all my music.also my ipod seems to keep resetting its self because the apple logo keeps showing i'm afraid i'm not very good with tecnology i remember going to my ipod on a disc management app and pressing benchmark not to sure what i did i think i nrrd to reinstall yhe softwear on my ipod but the apple home page keeps telling me to download the lates itunes,but as i have stated this is not possible.can anyone please help me?.ps out of intress i have another iopod the same model can i transfer the settings from one to the other.regards mr france

Sorry, I can only advice that you go to ubuntu ipod forum, for a better response.
Here I can suggest install iTunes and Restore the iPod.

Is there a way to force color management ON in Flash Player 10

Here's my problem:
I have a wide gamut display (calibrated and profiled) and
with a wide gamut display, it is very important to have web
browsing (and ideally everything else) fully color managed, because
unmanaged colors get displayed horribly wrong.
Firefox 3 supports full end-to-end color management (when
switched on). It even wisely assumes sRGB for CSS color values and
untagged images. But of course, Firefox 3 can't control Flash
rendering and this makes it impossible to view web pages that
contain Flash with correct (or even near-correct) colors.
So is there a way to force Flash Player 10's color management
ON by default? I know it is probably not a good idea in many cases,
but in my case, it would work a lot better than having if off by
default.
I'm using Mac OS X 10.5.5.

The website may alter its page display based on the browser ID that is transmitted to the site from your browser. Some browsers on the iPad like iCab allow you change the browser ID to something desktop-oriented. Safari does not provide this capability.

Calling a pkg from managed code and then using reflection to call a method from a script task

Hi we run 2012 std. I have some pretty good evidence that when I call my pkg from a .net service, a script component in that pkg fails when trying 2 use reflection to load and invoke our .net message history method. The exception is either on
the invoke or in the message history method. I suspect its in the method but will take additional steps 2 verify.
But when I run the pkg stand alone, it has no problem and the method does what it is supposed 2 do.
There r no vars passed from the service to the pkg. I wonder if its a managed to unmanaged to managed issue that the community is already aware of. If not, my apologies 4 posting this quickly.
I'll post more info here as I collect it.

we have 2 theories after showing the exception trace to folks who r more adept at managed code.
the first is related to the fact that our 3rd party dlls (I think entity framework is included in these) r older versions. I don't want to discount this theory but we have some evidence already that this might not be true.
I hope I can do justice to the 2nd theory but will make it clearer and clearer as I get a better understanding. I believe this is what Arthur was saying and I applaud his instincts. They explained that .net is so "smart" that it detected
a change in namespace (ie context as Arthur said) and purposely threw an exception 2 save us from ourselves. The workarounds discussed were a bit over my head but I will continue trying to better understand it. The fact that many of the methods
we call after reflection r now merged into one assembly seemed relevant to the discussion and possible workarounds.
this link came up in their discussion and I believe the bottom line here is that by qualifying assembly names further (in config?)r, a workaround is possible.
http://msdn.microsoft.com/en-us/library/system.type.assemblyqualifiedname(v=vs.110).aspx .
This link came up as well and has something to do with ILMerge and workarounds to ILMerge.
http://elegantcode.com/2011/04/02/dynamically-load-embedded-assemblies-because-ilmerge-appeared-to-be-out/ .
Finally, this link came up and seems to have something to do with embedding your dlls in one assembly without them losing their identity.
http://blogs.msdn.com/b/microsoft_press/archive/2010/02/03/jeffrey-richter-excerpt-2-from-clr-via-c-third-edition.aspx
I'll post more here as we muddle thru this.

HELP with sRGB in Photoshop Color Management Winows Vista

Greetings beautiful Windows people!
I am stumped with a Windows Vista problem — it is not displaying sRGB properly.
I posted bigger files here if this is easier to follow
http://www.gballard.net/windows_srgb/
But here goes:
First, my main question is HOW Windows Vista deals with sRGB because I am seeing inconsistant sRGB color on a client's workflow.
The first clue is:
How Photoshop is displaying Tagged sRGB on the two machines:
The original sRGB tagged file opens as above on my Mac.
Note: In the lower screen shots, Firefox has color-management enabled — and Photoshop's Working RGB is set to sRGB (it is using the sRGB embedded profile).
In the above Windows screenshot — Photoshop and Firefox look yellow — Explorer displays TAGGED sRGB correct — FireFox incorrectly matches Photoshop.
In the above Windows screenshot — Photoshop looks yellow — Explorer 8 and Firefox display UNTAGGED sRGB correct.
BECAUSE THIS IS CONTRARY TO THE THEORY, I suspected a bad profile so we loaded sRGB as the Monitor profile in the Windows Color Management setting.
Setting Windows Vista> Control Panel> Color Management: to sRGB fixed the problem (above, Photoshop is displaying the file correctly now, and sRGB is behaving as expected in both managed and unmanaged Web browsers).
However, I am clueless about two things:
Is Vista's "Color Management" Control Panel setting its Monitor RGB, or
Is Vista's "Color Management" Control Panel setting a DEFAULT RGB for the system?
I suspect the client has a bad hardware puck and/or buggy profiling package that needs an update because it appears to be even causing Photoshop to display sRGB improperly using the custom profile (and this is happening on eight of the client's Windows boxes).
He is using a glossy screen on the laptop I saw if his package can't handle it).
Can someone please clue me here as to what's going on (and how Vista should be set up to display sRGB consistantly in Photoshop and WEB BROWSERS)?

Thank you Chris,
When I did windows update I must have checked to download new display drivers I'm using (Dell with Dell Screens). It was loaded into the color settings as a profile, I hit delete and now all is well.
-Patrick

Your opinion's on managed switches? I need one

I'm looking at getting possibly a couple managed switches for our office. We have the need for multiple VLAN's and quite a few features now available in them as we grow! I've dealt with Netgear for a long time and find them to be pretty reliable especially with their ProSafe producs having a lifetime warranty.
Which is the best buy for a reasonable price?
http://www.netgear.com/Products/Switches/AdvancedSmartSwitches/GS748TR.aspx
this is what I was looking at, any other suggestions or opinions? thanks

I have bought 3Com, HP and LinkSys switches. Both managed and unmanaged. A couple of unmanaged HP switches have breaked down over the years, but haven't caused major trouble. The next switch I will buy will be a Cisco model. It has a configuration file in plain text, which is easy to compare to previous editions. It has all the features I can ask for. Maybe it does not have a fancy web based interface, but the command line is full featured. Linksys switches for example have only a limited set of instructions available over command line. Cisco has a separate console port, which again gives the admin full access to the setup. This is important when you have screwed up the configuration and cannot access the switch over the network.

System Center 2012 Endpoint Protection Deployment State of Client= unmanaged

Hello Everyone,
We are facing issues deploying SCEP on machines. The deployment state for clients shows as un-managed. Client settings & AntiMalware policy have been configured properly. SCCM Client has been installed successfully, however the Endpoint Protect deployment
state would not change to "To be Installed"/'Managed" from "unmanaged". Even after trying to install SCEP manually along with policy it shows as unmanaged. Please suggest how do i proceed further in this case.
Thanks
Sumit

Below is what it shows in EndpointProtectionAgent.log
=========
Handle EP Deployment policy. EndpointProtectionAgent 06/05/2014 15:53:48 4648 (0x1228)
Failed to get EP event code under registry key SOFTWARE\Microsoft\CCM\EPAgent EndpointProtectionAgent 06/05/2014 15:53:48 4648 (0x1228)
Failed to get EP event message under registry key SOFTWARE\Microsoft\CCM\EPAgent EndpointProtectionAgent 06/05/2014 15:53:48 4648 (0x1228)
start to send State Message with topic type = 2001, state id = 1, and error code = 0x00000000 EndpointProtectionAgent 06/05/2014 15:53:48 4648 (0x1228)
Start to send state message. EndpointProtectionAgent 06/05/2014 15:53:48 4648 (0x1228)
Send state message successfully EndpointProtectionAgent 06/05/2014 15:53:48 4648 (0x1228)
Service startup notification received EndpointProtectionAgent 06/05/2014 15:54:54 2504 (0x09C8)
Endpoint is triggered by CCMTask Execute. EndpointProtectionAgent 06/05/2014 15:54:54 2504 (0x09C8)
File C:\Windows\ccmsetup\SCEPInstall.exe version is 4.1.522.0. EndpointProtectionAgent 06/05/2014 15:54:54 2504 (0x09C8)
EP version 4.1.522.0 is already installed. EndpointProtectionAgent 06/05/2014 15:54:54 2504 (0x09C8)
Expected Version 4.1.522.0 is exactly same with installed version 4.1.522.0. EndpointProtectionAgent 06/05/2014 15:54:54 2504 (0x09C8)

Login stopped asking me if I wanted to have a managed login

When logging into our imacs, the login process used to ask me if I wanted to have my login managed or not managed. It has stopped doing this, and I would like to re-enable the option. Anyone know where this option is controlled? Thanks.

Hi David-
Your answer may explain why this question seems to baffle folks. The High School where I work has 18 new imacs. We had just enough money to buy the macs and pay an expert to help us set them up. We do not have a support plan, so that's why I can't go back to the experts and ask them...
When they set the computers, we authenticate users using our Active Directory server. Since I am a domain admin, when I log onto any of the computers I should have full priviledges. In the beginning when I logged in, the login screen would ask me if I wanted my login account to be managed or not managed. I think that's because, as a domain admin it would give me the option of testing the managed settings or just logging in without managed settings. By "managed" I am particularly talking about the System Preferences, where, for managed users, many of the items are grayed out. When unmanaged, I have access to all of the SysPref items.
So, the login screen that asked if I wanted "managed or unmanaged" also had a check box that said "Keep these settings from now on" or similar. I must have clicked that checkbox and then logged in as a managed user, so now I can no longer log into that machine as an unmanaged user.
I can still login as the local admin, so if there is a fix to my question, I should be able to address it.
Thanks,
Jeff

AccessViolationException - Connection.Open()

Hey guys, I have a really severe problem not beeing able to persist a suspended Windows Workflow Foundation workflow. (using Oracle.DataAccess 2.111.6.20)
I want to persist such a workflow to the database when it get´s inactive. My problem is in exact that case I am not able to open a OracleConnection to the database. If I open the connection before or after the actual persistence should take place everything works as expected.
The result on con.Open() are just these 2 exceptions:
A first chance exception of type 'System.AccessViolationException' occurred in Oracle.DataAccess.dll
A first chance exception of type 'Oracle.DataAccess.Client.OracleException' occurred in Oracle.DataAccess.dll
The testcode is kind of simple:
private void TestOraConnect()
// Create the connection object
OracleConnection con = new OracleConnection();
con.ConnectionString = "user id=test;password=pass;connection timeout=2147483647;data source=(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=myhost)(PORT=1521))(CONNECT_DATA=(SERVICE_NAME=myservice)))";
con.Open();
con.Close();
Thank you for any help, suggestions to overcome this problem.

ODP.NET consists of managed and unmanaged code. Every time a new ODP.NET version is installed, it inserts its bin directory onto the Windows PATH setting. Thus, all existing ODP.NET apps will then use this new installed version.
The problem that can arise is if you do not want all your ODP.NET instances using the same version. To support this scenario, you can specify the directory that each individual ODP.NET app should use to find its unmanaged DLLs via the DllPath setting in the Registry or .NET config file:
DLL search order discussed here:
http://docs.oracle.com/cd/E20434_01/doc/win.112/e23174/InstallODP.htm#sthref94
DllPath setting discussed here:
http://docs.oracle.com/cd/E20434_01/doc/win.112/e23174/featConfig.htm#BABCBFDJ

Stored Procedures - Get Column Information

Hello,
I recently switched from ODP Unmanaged to ODP Managed (I'm using Entity Framework).
The Unmanaged drivers were working fine after adding the necessary information in the web.config section. I could add the stored procedures and generate the complex types using the Function Import - Get Column information.
After the switch the config section was updated and it all appeared to work. By the way the changes in the format of the config-section is poorly documented - to say the least.
However when I try to generate the complex types again I just get a "System.notSupportedException" (Message: The specified type is not supported by this selector).
Was this functionality not ported to the Managed Drivers? Is there something that can be done to enable this functionality?

The implicit ref cursor config file format is different between Unmanaged ODP.NET and Managed ODP.NET.
To save yourself from pulling your hair out, do the following -
1) Install ODT 12.1 if you haven't already
2) Find the stored procedure in server explorer, right click it and run it, and enter input parameters.
3) For the output ref cursor that represents the return value for your Entity Function, choose "Add to Config" checkbox.
4) Then select either "Show Config" (and then cut and paste) or "Add to Config".
Here is a screenshot of what I am talking about:
http://i.imgur.com/t1BfmUP.gif

Handle leaks?

Has anyone seen handle leaks in ODP.NET? I have a multi-threaded server application that contains managed and unmanaged C++ code and uses ODP.NET. It leaks handles that seem to be thread handles created by ODP.NET. I cannot reproduce it in a simple test. I am working on it, but it would be nice to know if anyone noticed the same problem and has a workaround or fix. This is a big problem in my app because the number of handles goes up to 10,000 in a couple of hours. GC.Collect does not help.
Thanks,
Aleksey

Yes, I know that. I do close and dispose everything. The behaviour is hard to reproduce, but it does happen.
One quick tip you can try is to explicitly close all
objects you created when no longer needed. For
example, call the Close/Dispose method on your your
ODP.NET connections, DataReaders, etc. objects before
you leave your applications scope. Sometimes the
.NET garbage collector cannot reclaim resources
quickly enough under heavily-loaded conditions.
Alex

Managed vs Unmanaged ODP benchmarks

Similar Messages

Maybe you are looking for