Only Durability Required

Hi Everyone,
I wanted to know what's the best option when the requirement is only
Recovery while using Berkeley.
What i mean is I don't require the Atomicity, Concurrency properties of
a transaction but only durability.
To achieve good performance can i disable locking and other subsystems
considered while transaction handling.
Or is there a way so that we can achieve
Durability (App crash and OS crash) without transactions
Best Regards,
Mohd Aehthesham.

Hi Mohd,
We can discuss a lot about performance tuning... when we don't know the environment details.
Basically you are asking if you can replace transactions with a homemade solution, right? If you are experienced with Berkeley DB, you might know that performance is one of the most powerful characteristics of the application, if not the best. You can configure Berkeley DB in many ways, in order to improve your application performance, so this is why I think that is almost impossible to obtain a significant improvement from the performance point of view with a certain application, by writing your own recovery system, if Berkeley DB is configured accordingly.
First of all I think that you should see if using Berkeley DB with transactions will be, or not, good enough for your requirements. If not, you should be sure that the tuning is well done.
Transaction tuning: http://www.sleepycat.com/docs/ref/transapp/tune.html
Transaction throughput: http://www.sleepycat.com/docs/ref/transapp/throughput.html
Tuning on Java Edition (I think that you might want to take a look over Java Edition, which is having some different expectations and goals): http://www.sleepycat.com/jedocs/TransactionGettingStarted/perftune-intro.html
It's hard to establish an independent performance goal because the range of application characteristics is too big, and we are not having any clue about your application/requirements.
Anyway, if you want to build a new module, please be aware that the major performance test variations are:
- in cache/out of cache
- single threaded/multithreaded
- insert sorted, ascending data, or insert randomly generated (unordered) data
- use transactions/don't use transactions
Please, try to run the application using Berkley DB, and if you are experiencing problems with performance, follow the tuning hints or provide us your application characteristics and your performance expectations, in order to discuss on a real basis.
If there is anyone who tried this out, maybe he can share from his experience.
Regards,
Bogdan Coman, Oracle

Similar Messages

Can we display only the required dimensions under Dimensions Drop Down list

Hi,
I am using Hyperion Financial Data Quality Management, Fusion Edition 11.1.1.3.00, so far under Activities -> Maps option we used to display all the dimensions used in the application, but now as per the client's requirement we need to display only those dimensions which needs mapping under dimensions drop down list.
Eg)
Now in our Application our Dimension Drop Down list will look show these dimensions
Account
Entity
Legal Entity
Inter company
Reverse 1
Reverse 2
Among these we are doing mapping only for Account, Entity dimensions so we need to hide rest of the dimensions from the Dimensions Drop Down list for which we are not using any mappings (Legal Entity, Inter company,Reverse 1,Reverse 2)
Another thing is these dimensions should be hidden only from the drop down list and not from the Output file generated after the Export.
Some suggestions regarding this issue will be greatly helpful.
Thanks,
Abdulla Javeed Hassan

Hi
If hiding the other dimensions is to prevent users from accessing and changing the mappings then you could do this by amending the object security so that only certain user roles can access the Activities > Maps option.
This will only really work if you have limited users / central team that will control the maps but it is the most obvious way to stop users accessing the maps.
Hope this helps
Stuart

Resource Availability - for next Shift only when required

Hi Gurus,
How the Resources can be made available for the 2nd/3rd shift, only during a certain period. Resource should be made available for 2nd/3rd Shift as and when required period
Since the requirement for the shift is not sure during the Master Data(Resource) creation or at the Planning phase, It can only be decided upon during the shop floor execution based on the load and the urgency of the production compeletion. Is there a way to select/extend the Shift for 2nd/3rd shifts to make the resource available, ONLY as and when required
Appreciate your help in advance
Regards,
TamNen

TamNem,
1. How this Resource availability - Extension will get accounted in CRP
I am not sure why you even care. This is execution, not planning (predicting the future). In effect, you are declaring, either upon execution or very near to the actual execution date, that the existing plan was unworkable. You declared a new cap plan, based on the reality of the capacity situation. Many companies don't even bother with such niceties.
Anyhow, these additional hours will show as available in CRP. If you schedule your existing orders to utilize these hours, then the additional hours will be consumed, by any orders that are scheduled to consume them.
When the Shop floor user decides in the mid-of-the day and plans to run the 2nd shift for the production completion for only one day to achieve the target, should the same procedure Master data change be done(each a day - when required)
Who even cares? You will confirm orders during execution, based on what actually happened, and these confirmations will consume existing capacity requirements. Once you have confirmed your orders. why do you need to change your old Cap plan to match the consumption actuality that happened during execution? It is all a moot point, since it has already happened. Planning is important only to predict the future. What is the point of changing the old (inaccurate) plan to match the reality of what actually happened?
3. Since the production department has many number of machines(15 Machines) and all...
what are you talking about? I definitely don't understand the question.
Capacity plan lets you plan what you are going to do in the future. Execution tells you what actually happened. Who cares what machines you actually used/didn't use? You either executed the cap plan, or you didn't. Execution will seldom match the plan. The only thing a planner is interested in is "how much available capacity do I have now, and how much remaining load is there on the capacity, as of right now. No other questions are relevant. Previous plans are irrelevant. Comsumption variances are irrelevant (except to help you fine-tune existing future plans).
I don't believe there is much benefit to 'fiddling' with the previously created plan so as to make it match reality. Just execute. If manufacturing has to work additional shifts, let them do so. I don't believe It is really necessary to change the old plan to reflect the fact that manufacturing decided they needed to work additional shifts to meet the requirements of the cap plan..
Rgds,
DB49

Read-only filesystem required management...

I have had a netbook with Arch Linux on it, and have used it at school and home without problems. This morning at home, I rebooted when I meant to poweroff. So I used the button to poweroff while it was booting back up.
At school, when I turned it on, I was told to enter the root password, so that I could manage/fix my file-system. It also told me my file-system was read-only, and I had to remount it. I rebooted my computer numerously through the day, but kept getting the same thing. [I'm too much of a newb to fix something like this without the internet.]
Now, when I just got home to get to fixing it, it worked/booted up fine (albeit with a forced file-system check). So, there's no real problem, I'm just concerned. I'm supposing this happened because I did a hard power off at a bad time?

If you have some time, read up on the linux boot process.
I will give a brief explanation, however I can't guarantee that the explanation is 100% correct. As you likely know the first step in the boot process is having the mbr load (grub) and then loading the kernel from a hard drive, flash drive, etc... Once the kernel loads the kernel needs to mount the root file system. The option 'ro' in grub.conf tells the kernel to mount the root file-system as read-only. Once this is done the boot process continues and then later on will remount root and mount your other filesystems as it is configured in etc fstab.
It is hard to say what was wrong with your system has you didn't provide any of the errors. Likely the fs had errors on it and it was having a hard time remount as rw.
If I was in your shoes I would have entered the root password and ran a fsck on the disk in question.
Hope this helps clarify things a bit.

How to remove unnecessary data and print only the required data in JSP

Hi all
The following output is from TCP/IP connection of RFID Reader.
Connected to: /192.168.0.3
EVT:DCE DCEVERSION 3.2.0.0
EVT:DCE READER IM5 RFID Reader Ver 3.29
EVT:DCE READERINFO Basic Reader Interface Version Q
EVT:DCE READERINFO FCC 915Mhz
EVT:DCE READERINFO Copyright (C) 2002-2006 Intermec Technologies Corp.
EVT:DCE FIXEDATTRIB CHKSUM=OFF
EVT:DCE FIXEDATTRIB TTY=OFF
EVT:DCE FIXEDATTRIB ECHO=OFF
OK>
EVT:TAG H02000031323334000151E62B 1 1
EVT:TAG H02000031323334000151B775 1 1
For my JSP case I have converted above output's program in javabean I want to print only tagID and other values in a table in my JSP page. Could anybody let me know how to ignore the unnecessary data and print only values after EVT:TAG in three columns and rows, besically rows are not fixed because it depends on reader how many tags it detects?
Thanx
kvijai

Hi,
Try this
private function doFilter(item:Object):Boolean
     var searchString:String = myTextIp.text.toLowerCase();
     if (item.name.search(searchString) == 0)
     return true;
          else
     return false

IE 10 blank new tab and missing favourites problem only when requiring a proxy for first page visit

We are having a very strange situation with one (only) of our 64bit 2008r2 remote desktop session host servers. The problem we are having is that when any new tab or window is opened it must first open a page that is in the
internet options -> connections ->LAN settings -> Advanced -> Exception. Do not use proxy server for addresses beginning with
Area. OR ELSE....
The entire contents of the favourites menu below "Organize favourites" is missing, though the favourites explorer bar, and tool bar display the links, but they cannot be opened in the current tab and will not create a new tab when you click open
in new tab. HOWEVER, this problem is instantly resolved if you manually type into the address bar and visit a site that is in the proxy exception list. Then that tab starts to function normally, the favourites menu is populated and works, and so
does the favourites explorer bar, and toolbar, and other websites can be visited.
This is extremely problematic. Looking at the proxy server logs, when the problem is happening no attempt is made at contacting the proxy server.
Ways we've encountered this problem
If the user has the option set that a new tab opens the 'new tab page' then any new tab experiences the problem.
If the browser is set to open the user's home page(s) then any of the home pages that are not in the exception list may or may not open successfully. But again any of them that opened a page that was not in the proxy exception list will experience
the problem
when a user tries to click on a website link from an email or document that is not in the proxy exception list, as this opens a new window or tab and it is just blank.
when a website not in the exception list tries to open a new window/tab
This problem is happening for all users on that RD session host server, including domain admins, except for THE domain/enterprise administrator. Our other 2008r2 64bit RD session host servers and windows 7 machines with the same version of IE
installed on it does not have this problem.
We have tried deleting all IE data, resetting IE, re-registering the ieproxy.dll with the regserv32 command and the problem continues.
Our proxy settings for each user is coming from the group policy setting of 'computer config -> policies -> Administrative Templates -> Windows Components -> Internet Explorer -> Make proxy settings per-machine = enabled'
and THE domain administrator choosing to enable the proxy server and disable automatic proxy detection.

Hi,
Since you have tried many methods like deleting all IE data, resetting IE, and it works well on other Windows 7 machines, the problem might be some clients didn’t respond to the group policy.
Try to update the group policy with this command at the clients ”gpupdate /force”.
And check whether the client is infected with virus, some virus might cause this issue too.
Regards
v-yamliu

Are IWBK only considering requirements on the component screen on PM orders

Hello
We use External operations on PM orders. And from that are created purchase requisitions?
But it seems that this purchase requisitions is not validated by IWBK?
If I create exactly the same purchase requisition from the component screen on the PM order, IWBK is validation it?
Is this how IWBK shall work?
Best regards
Tom Saga

Hi,
IWBK includes only materials that are specificed in the components tab. External operations are usually work tasks that are hired in and are therefore not relevant for an availability check.
-Paul

Can we display only the required columns dynamically?

Hi All,
I have a scenario where in I have 30 columns in the rtf template. All the columns will never be displayed, only a select few will be available for display. So the output file is show blank columns. Now I want to display only the columns which contain data.
FYI- We cannot know which columns consist data until we run the report.
Any Advice.
Thanks in advance.
Madhav.

Hi,
You can use the following:
Column header:
<?if@column:/node/child_node="PUBLIC"?>My label<?end if?>
Column data:
<?if:/node/child_node="PUBLIC"?>My data<?end if?>
The @column will tell XMLP wheater or not the column shall be shown.
I hope this helps you.
BR Kenneth

OAS 10g (10.1.2) - Read only Access Required to oc4j Container Settings

As a load tester I need to be able to access the details of the container settings for our live servers to enable me to mimic this set up on my test kit.
of particular interest are the Server Properties and Data Sources for each container.
I am not permitted a username/login for the live OAS.
The Unix admins looking after this kit tell me it's not possible to grant me read only access.
Is this the case?
Is there any other way for me to get this information without a live login - can OAS export the details of these settings?
It's very frustrating to have to rely on info handed down from the Unix admins as and when they have the time to give it to me!....
Any help much appreciated....

The Unix admins looking after this kit tell me it's not possible to grant me read only access.
Is this the case?
NO, the UNIX Admins are full of it. Here are several (but not all settings)
400 Owner Readable
600 Owner Readable and Writeable
700 Owner Readable and Writeable and Execute
740 Owner Readable and Writeable and Execute Plus Group Readable
760 Owner Readable and Writeable and Execute Plus Group Readable and Writeable
770 Owner Readable and Writeable and Execute Plus Group Readable and Writeable and Execute
774 Owner Readable and Writeable and Execute Plus Group Readable and Writeable and Execute and Other Readable
776 Owner Readable and Writeable and Execute Plus Group Readable and Writeable and Execute and Other Readable and Writeable
777 Owner Readable and Writeable and Execute Plus Group Readable and Writeable and Execute and Other Readable and Writeable and Execute
Is there any other way for me to get this information without a live login - can OAS export the details of these settings?
Sorry I don't know the answer to this one
And don't tell the Unix admins your handle is "TestCowboy" most of them don't have much of a sense of humor. (But I think its great!)

Propogating Only the required data to destinition

We are planning to use replication using streams.
We have following scenario
'Site M' is a master site
'Site A' and 'Site B' are the secondary sites.
The updations on 'Site M' are to be propogated to site A/Site B.
Only the records which are ment for site A should be
PROPOGATED to site A and
Only the records which are ment for site B should be
PROPOGATED to site B.
How do we do this.
We just know one way that using row subsetting at apply process level
at destinition site (dml_condition) or using dml_handler (again at destinition site) we camn achieve this.
But this has a problem of transferring the unnecessary data to destinition.
So we are looking for an approach which will allow us to transfer only the needed data to the destinition site and hence reduce the network traffic. How do we doit.
Is there a way whereby we can do it tgrough PROPOGATION process ... or
any other way.

Rules define what is captured, propagated, and applied with Oracle Streams. You can write your own custom rules for propagation that will limit the propagated changes to a particular subset to simulate the effect of ADD_SUBSET_RULES. See chapter 6 in the Oracle Streams Manual (9.2.0.2) available at
http://download.oracle.com/docs/cd/B10501_01/server.920/a96571/strm_rules.htm#49200
In the section on System Generated Rules, there is a discussion of the rules generated by ADD_SUBSET_RULES. You will need to write rules that are similar to these. In addition, you will need to assign a rule-based transformation to the custom rules so that the LCR can be manipulated appropriately. For example, an UPDATE at the source database might need to result in a DELETE at one apply site and an INSERT at the other apply site. You will need to write a transformation on the propagation rule where the DELETE should occur to:
a) change the command_type to DELETE
b) remove the NEW values from the LCR.
Similary for the apply site where the INSERT should occur, the transformation on the propagation should
a) change the command_type to INSERT
b) move the OLD values to NEW values
You can simplify the rules shown in the documentation by creatin[i]Long postings are being truncated to ~1 kB at this time.

Need help making digital signature field read only if required fields are null

Using Adobe LiveCycle Designer, I have a digital signature field. I want to prevent users from being able to digitally sign the pdf if any required fields in the pdf are null or blank. I'm a newbie to scripting so any detailed help would be greatly appreciated. The required fields in the pdf include radio buttons, check boxes, text field, etc. I appreciate any help anyone can provide.
Thanks

Exactly how you check a field for being empty depends on what type of field it is. For text fields, you should get the field's value as a string and compare it to an empty string, something like:
var val = getField("Text1").valueAsString;
if (val) {
app.alert("Field is not empty");
If there is a default value that needs to be changed, you would compare to the default value:
var f = getField("Text1");
var val = f.valueAsString;
if (val !== f.defaultValue) {
app.alert("Field value is OK");
For check boxes and radio buttons, you'd compare to the string "Off", which means none in the group is selected.
For dropdowns (combo boxes) and list boxes, there is normally one value selected, so you would have to get the value and check against the default value (or something else) to see if the field is properly filled in.
You could place the script there, but I usually use a separate button that I encourage the user to click in order to check the field. The problem with using the Will Save event is you can't prevent the save from happening.
The script can loop through all of the fields (and test all required ones), or you could set up an array of field names to check.

How to print a report without border, only data required

Hi,
I have a report and need to print it. While printing I don't want the borders of the report in the output. Is that possible?
Please let me know.

There is a printer friendly page template that may assist you. This is the final parameter in your URL.
http://docs.oracle.com/cd/E23903_01/doc/doc.41/e21674/concept_url.htm#BEIFCDGF

FlashBuilder integration - compiling only the required test

We have a large project and I'm looking into the best way of running Unit Tests on a specific class, without having to recompile the entire project. Is this possible with the FlashBuilder Premium integration?

Interesting, this looks like it could be the way forward. How does the task know which SDK to compile with, and what about more advanced compilation options (like
define, which we use occasionally for conditional compilation).
And, on the general subject, would it be possible to start FlashBuilder debugging from the ant script?

Listing a join column only once in a select

Hi. Has there been any advancement in t-sql where I can continue to select * on tables a and b in a join but expect to get back only one occurrence of columns that r used in the join. Or do I still need to explicitly list all cols in the
query?
eg. select a.*,b.* from a inner join b on a.c = b.c
hoping there is some shorthand way to see c only once

Sorry, I would not refer your expectation with "advancement". Relational engine needs to identify each column separately in the query, user have capability to select required columns.
If you consider LEFT OUTER JOIN, how would it decide which column to display from LEFT or RIGHT table (because when the values do not match you will still have all the rows from LEFT table).
When you specify "*" it brings the entire result set. So you will have to do little work and select only the required column. :-)
If this post answers your query, please click "Mark As Answer" or "Vote as Helpful".

What are the bare minimum required tags for DNG (digital negative) images.

I'm trying to write a program that converts a standard BMP file into DNG. I know that isn't practical for many things, but I'm curious as to what the minimum required tags are for a DNG image. The DNG specs at http://www.adobe.com/content/dam/Adobe/en/products/photoshop/pdfs/dng_spec.pdf show what tags are available, and mention whether or not a given tag is required, but this document doesn't have a list showing only the required tags. So I'm wondering, does someone here have a list of only the required tags for DNG images?

Just discovered there's a dedicated DNG section to the forum at Digital Negative (DNG) which is where this question should probably be, but don't know if think I can move this message myself. Maybe a moderator can move this post.

Only Durability Required

Similar Messages

Maybe you are looking for