Replacing natural keys with surrogate keys

Hi,
We have database that is runnig with some old applications since many years. In the database, all the primary key are natural keys (composed with many columns). The applications are developped with mod pl/sql and oracle forms6i / designer6i.
We are planning to develop a new system (replacing all the old applications), the database tables will ramain the same but we are thinking about replacing the natural keys with surrogate keys. Since we are planning to develop and migrate into the new system gradually (oracle forms applications will not be redeveloped first), my question is:
- will the oracle forms applications still works?
- what is the risk and precautions to take ?
- is the decision to convert natural keys into surrogate a good one?
All your comments and propositions are welcome.
Thanks.

Surrogate keys in general is a good idea. Many generators cannot handle multiple column PK's very well, or not at all. I believe, for instance, that Apex can handle 2 columns only.
But, doing this on an existing datamodel can be a lot of work. You have to change your tables, of course, but also the foreign keys, triggers, constraints etc. A PK is protected from an update, but I guess you also want your old, natural PK, to be protected as well.
Forms generates master-detail synchronization based on FK relations. If these change, you may have to change your Forms too. As long as you don't do anything in your existing form, my guess is that it will still work. The forms will maintain the FK relation between old table definitions. This works independently of the actual FK relation in the database. You can create triggers on the tables to actually populate the new PK and FK.
For something like this: give it a try in a small test setup first. Than calculate the risk and impact.

Similar Messages

Loading data into Fact/Cube with surrogate keys from SCD2

We have created 2 dimensions, CUSTOMER & PRODUCT with surrogate keys to reflect SCD Type 2.
We now have the transactional data that we need to load.
The data has a customer id that relates to the natural key of the customer dimension and a product id that relates to the natural key of the product dimension.
Can anyone point us in the direction of some documentation that explains the steps necessary to populate our fact table with the appropriate surrgoate key?
We assume that we need to have an lookup table between the current version of the customer and the incoming transaction data - but not sure how to go about this.
Thanks in advance for your help.
Laura

Hi Laura
There is another way to handling SCD and changing Facts. This is to use a different table for the history. Let me explain.
The standard approach has these three steps:
1. Determine if a change has occurred
2. End Date the existing record
3. Insert a new record into the same table with a new Start Date and dummy End Date, using a new surrogate key
The modified approach also has three steps:
1. Determine if a change has occurred
2. Copy the existing record to a history table, setting the appropriate End Date en route
3. Update the existing record with the changed information giving the record a new Start Date, but retaining the original surrogate key
What else do you need to do?
The current table can use the surrogate key as the primary key with the natural key being being a unique key. The history table has the surrogate key and the end date in the primary key, with a unique key on the natural key and the end date. For end user queries which in more than 90% of the time go against current data this method is much faster because only current records are in the main table and no filters are needed on dates. If a user wants to query history and current combined then a view which uses a union of the main and historical data can be used. One more thing to note is that if you adopt this approach for your dimension tables then they always keep the same surrogate key for the index. This means that if you follow a strict Kimball approach to make the primary key of the fact table be a composite key made up of the foreign keys from each dimension, you NEVER have to rework this primary key. It always points to the correct dimension, thereby eliminating the need for a surrogate key on the fact table!
I am using this technique to great effect in my current contract and performance is excellent. The splitter at the end of the map splits the data into three sets. Set one is for an insert into the main table when there is no match on the natural key. Set two is when there is a match on the natural key and the delta comparison has determined that a change occurred. In this case the current row needs to be copied into history, setting the End Date to the system date en route. Set three is also when there is a match on the natural key and the delta comparison has determined that a change occurred. In this case the main record is simply updated with the Start Date being reset to the system date.
By the way, I intend to put a white paper together on this approach if anyone is interested.
Hope this helps
Regards
Michael

Best Practice loading Dimension Table with Surrogate Keys for Levels

Hi Experts,
how would you load an Oracle dimension table with a hierarchy of at least 5 levels with surrogate keys in each level and a unique dimension key for the dimension table.
With OWB it is an integrated feature to use surrogate keys in every level of a hierarchy. You don't have to care about
the parent child relation. The load process of the mapping generates the right keys and cares about the relation between the parent and child inside the dimension key.
I tried to use one interface per Level and created a surrogate key with a native Oracle sequence.
After that I put all the interfaces in to one big Interface with a union data set per level and added look ups for the right parent child relation.
I think it is a bit too complicated making the interface like that.
I will be more than happy for any suggestions? Thank you in advance!
negib
Edited by: nmarhoul on Jun 14, 2012 2:26 AM

Hi,
I do like the level keys feature of OWB - It makes aggregate tables very easy to implement if your sticking with a star schema.
Sadly there is nothing off the shelf with the built in knowledge modules with ODI , It doesnt support creating dimension objects in the database by default but there is nothing stopping you coding up your own knowledge module (use flex fields maybe on the datastore to tag column attributes as needed)
Your approach is what I would have done, possibly use a view (if you dont mind having it external to ODI) to make the interface simpler.

[SOLVED] Can't sign key with pacman-key

I've been reading and digging all over the place but I can't seem to find anything on this.
1. I have created my own keys for signing packages that I want to put in a local repo so I don't have to compile the same package over and over on all my machines.
2. I have imported said key using sudo pacman-key --add rdjack21.pgp
3. Used sudo pacman-key --lsign rdjack21 I get this output:
-> Locally signing key rdjack21...
==> ERROR: rdjack21 could not be locally signed.
4. Ok so lets sudo pacman-key --edit rdjack21 output:
gpg> sign
pub rsa2048/3D1B5544
created: 2014-11-30 expires: never usage: SC
trust: ultimate validity: ultimate
Primary key fingerprint: B0D7 C728 31FA B2B0 6A3C 7086 DC7F DE0C 3D1B 5544
Richard Jackson <[email protected]>
Are you sure that you want to sign this key with your
key "Pacman Keyring Master Key <pacman@localhost>" (645ECB0E)
Really sign? (y/N) y
gpg: signing failed: No pinentry
gpg: signing failed: No pinentry
gpg>
5. To hack this and make pacman work with my local repo signed by the rdjack21 key I had to change the trust to ultimate but I do not like that solution. I would prefer to sign the key and move the trust back down to full which is where it should be.
This is happening on all of my boxes not just one of them. I did find something on the web about "No pinentry" message above and the sugestion was to change the owner of the tty I'm using to be root for the signing to work but that did not work for me (su'ed to root first instead of running everything with sudo).
While changeing the trust works that just feels wrong to me. I really should beable to sign the key.
Last edited by rdjack21 (2014-12-01 01:50:05)

Taking a clue post gnupg 2.1.0-4 upgrade breaks passphrase via STDIN I was able to solve the problem and I can now sign keys.
The solution is add "pinentry-mode loopback" to /etc/pacman.d/gnupg/gpg.conf
And create the file /etc/pacman.d/gnupg/gpg-agent.conf and put "allow-loopback-pinentry" in the file.
Once that is done everythings works as expected!
Marking this as solved.

Problems Engineering Surrogate Primary Key with Unique Key

SDDM 3.3.0.747 with 2 problems (at least so far). I am hoping that the problem is with this SDDM rookie and I have overlooked some setting. PROBLEM 1 I don’t want to start a religious debate about surrogate vs. natural keys but I am having a problem engineering both correctly from the logical model. I am a rookie when it comes to SDDM but have many years of experience with Designer. By default I like to have both a natural UID (UK) and a surrogate key based primary UID (PK) which is used for foreign keys. The problem I am having with engineering is I can successfully engineer the surrogate PK’s, engineer the FK’s using the PK’s but cannot get the unique key to contain the surrogate keys in the child table. If I check the identifying property in the relations, the PK columns and the UK columns are included in the child PK and the UK contains no columns. The Setup I have defined two reference entities, PROBABILITY and SEVERITY with natural unique keys defined. I also have a child entity RISK_ASSESMENT with relationships back to the PROBABILITY and SEVERITY entities and both have the “Use surrogate keys:”: check box checked. The unique key for the RISK_ASSESMENT entity includes the relationships back to PROBILITY and SEVERITY. None of the entities have a PK or surrogate key defined and they all have the “Create Surrogate Key” check box checked. In addition the following preferences are set: Data Modeler/Model/Logical NOT Checked - Use And Set First Unique Key As Primary Key NOT Checked – Name – Keep as the name of the Originating attribute Checked – Entity Create Surrogate Key Checked – Relationship Use Surrogate Key PROBLEM 2 When the foreign key columns are engineered I want the names to have a prefix “FK_” but they don’t. Templates are set as follows: Data Modeler/Naming Standard/Templates Foreign Key: FK_{child}{parent} Column Foreign Key: FK_{ref column} Engineer to Relational Model/General Options Checked - Apply name translation Marcus Bacon

I have been switching between SD 4 EA1 and SDDM 3.3 trying to get things to work and trying out the template table for adding audit columns (really nice!).
Concerning Problem1. No matter what settings I use and whether I use SDDM 3.3 or SDI cannot get the FK columns to be included in the UK even though the relations are included in the UID in the entitty. When I open the properties of the child table and click on the naming standards button and click ok it complains that the UK is not complete. I add the FK columns to the UK and all is well including the naming standards.
Concerning Problem 2. Sometimes it engineers the names for FK's from the template and sometimes it doesn't. Didn't see a pattern. Gave up trying and used Naming Standards button. I still had to change a few.
The good new is, that after make changes needed in UK's and Column names of 18 tables, I know have everything deployed to Test except FK Indexes. I think I have to do those by hand.
Marcus Bacon

Key probelm w. Generic EnumMap -- How to replace a value with unknown key?

Hi, i have a problem that i want to replace an value of an key/value pair, but I dont have the approiate key for that, to avoid passing the key throug all methods, I just implemented
a method in the delegate class.
public K getFirstAssociatedKey(BilanzObjekt bo){
           Set<K> keys = this.container.keySet();
           for(K key: keys){
               V value = this.container.get(key);
               if (bo == value){
                    return key;
           return null;
     }It works, but I can't use the retrieved key for a put-statement, cause the satic type comparision seems to fail. Even the key is retrieved from the original, the compilier is unable to validate that the key is in the approiate Enum, so it refuses the operation.
In my special case I can avoid this by returning the value to the calling mehod inside the delegate, but that can't be the solution for all cases. Even If i would pass the key-vaue through all methods, the universial declaraction of ? extends Enum<?> would raise the same problem, that the compiler can't verify if the given key ist part of the Enum of the current EnumMap.
Cause i still not an expert dfor generics, I might have overseen an solution.
So is there any way to retrieve valid keys for the put-method at runtime without declaring a specilized subtype? Using rawtypes would supress compiler-errors, but that isn't the way what Generics intends. I allready have some lager Enums and some subsuets with EnumSet. But this solution isn't absolutly clean, cause you have to use the EnumSet to Iterate over the map and can't use keySet() anymore (Without special preparations for the values not contained in the EnumSet). Also I don't want to limit the code to only one type of Enum. So I my looking for another solution.
Is there hoepfully any easy and gernerally solution to that problem?
Thanks a lot in advance.
Greetings Michael

Hi, I redesigned my programm a little bit by returning the Objeckt which should be replaced to the delegate, but this causes the problem that I know that it is the exact type, cause the previous value had the same type before, but I can't replace it as long I don't except a fixed type for the value-parameter of the EnumMap.
The workaround which worked with keys via Map.Entry, cause I didn't replaced them, don't work here.
I found a solution which works for me (after the redisign mentioned above), even it is quite dirty., cause it moves the error detection form compiletime to runtime. But to gain more flexibilty, i prefer this. The else clause is just to find some programming faults as early as possible. I was quite suprised that the cast to V is legal, but the put -method will throw a ClassCastEcxception if it is not. So its a trade off betwenn flexibility and early error detection. I'm sure that should not used widely, but is a soluion or workaround for my problem.
/* dirty cast */
V new_v = (V) old.getReplacingValue();
                    if ( new_v != null){
                              /* If the both instance are of the same base class, this operation is safe,
                               * Cause if the previous value was ok, then the new value must be also OK!
                               * This should be the default case.*/
                         if(new_v.getClass().getName().equals( old.getClass().getName()) ){
                                     this.container.put(key, new_v);
                           }else{
                         /* Otherwise a class-cast exception is most likely,
                          * but should normally not reached, if replacing methods are always returning the most specialized type and are overwritten in all subtypes. */
                             try{
                              this.container.put(key, eBoNew);
                              }catch(ClassCastException cce){
                              cce.printStackTrace();
               }Greetings Michael

Replace Temporary License Key with permanent Key

Hi Folks,
I have installed BOBJ and installed SAP integration Key on VMW with temporary SAP Integration kit Key. Integration is working fine and all the roles are imported from SAP BI to BOBJ.
Now this Key has to expire after few days. I am to get permanent key for this SAP integration Kit.
How can I replace this old temp key with the new key?
Thanks,
Bashir

Hi,
as mentioned above - for the server components >> BOE - you need to go to the Central Management console.
For the client components such as Crystal Reports or the SAP Integration Kit you need to use the License manager
Ingo

How to use Control Key with other key as Ctrl+C

I am using this code to access Shift key + C
addKeyListener(new KeyAdapter()
public void keyPressed(KeyEvent KEv)
if(KEv.isShiftDown() && (KEv.getKeyCode() == KEv.VK_C))
//Shift + C key down
Please tell me how to do with Control Key?
Vishwajeet

addKeyListener(new KeyAdapter()
public void keyPressed(KeyEvent KEv)
if(KEv.isControlDown && (KEv.getKeyCode() == KEv.VK_C))
//Cntrl + C key down
//if u want to create a key stroke for keys with contrl combination use CTRL_MASK.

Re: CS5.5 Keying with Ultra Key not working like the tutorial.

I was watching the tutorial about green screening. It looks really good when I was doing it but when I played the whole video there were some shadows still. Do I have to find the darkest part of the video to make it all green screen properly?

Try it.
You may also need to apply more than one key effect to the same clip.
You may also need to use all the tweaks in the effect to get a great key.

Surrogate Keys vs. Natural Keys

Hi All,
Is anyone aware of a recommendation regarding the use of surrogate keys vs. natural keys?
Regards,
Irfan Abdul Rehman

The Natural Keys approach was first. This was the approach used when Relational Databases were first discovered. But I believe, it was based on the premise that the design of the Database does not change over time. Most people seem to side with one or the other. People's opinions here are usually based on what they were brought up with or personal experiences where they have run into problems with one approach or the other.
When viewed within the context of object-oriented design, the Surrogate approach is more common. Consider: does the uniqueness of a table have any relevance to other tables? If the uniqueness of a table changes, why should this have to impact other tables? If a customer is associated with an Order, using Natural Keys, you need to know what columns make the Customer unique when inserting a record into the ORDERS table. With Surrogate Keys, you don't need to know what columns make the Customer unique, the customer is referenced by its Surrogate Key. With Surrogate Keys, you only need to know what makes a Customer unique when dealing with the CUSTOMERS table.
With SQL Server, the Surrogate approach is very straight forward. In Oracle, the appeal of Surrogate Keys is less than with SQL Server as there is a more work to implement them. In SQL Server you specify the Column as an identity column. In Oracle you need to additionally add a sequence. In addition, if you want the value automatic on inserts, you will need to create a trigger (unfortunately here, when you insert, there is no way to find out what you just inserted). If you don't choose to have the value automatic on inserts, your insert SQL statements will require extra SQL code and the Surrogate value is not enforced (i.e. someone could enter any value and this could lead to a Key Violation)
Here are some disadvantages of Natural Keys:
-     Almost always, more columns to join on. If Table B is a detail table of master A and C is a detail to B and D is a detail to C, you will need at least 3 columns to join D to C in an SQL query.
-     If your uniqueness of a table changes (Ex: the number of Columns making your Table unique changes from 2 to 3), with Natural Keys, all of your SQL (Stored Procedures, Reports, Views, SQL Scripts, Application Code) will have to be re-written and your foreign keys relating to that table will have to be changed. With Surrogate Keys usually as simple as modifying that table.
-     If the data type of a Primary Key column changes (Ex: you used a varchar(20), now it's not big enough and has to be changed to varchar(100)), with Natural Keys, all Foreign Keys related to that Table will have to be changed (may also impact SQL Code). With Surrogate Keys usually as simple as modifying that table.
-     The Classic: a value changes in the Primary Key (Like Last Name). Now you've got to update that in every Foreign Key. Which means you'll have a big headache when you have to temporarily drop the constraint(s).

Surrogate vs. intelligent keys with adf/bc

i'm currently using intelligent keys to define a primary key on my tables. I'm considering changing this approach and using surrogate keys instead. Consider the current tables and their relationship:
Application (table's name)
application_name (primary key)
Menu (table name)
application_name (part of composite primary key, also the foreign key referencing application_name in the application table)
menu_name (part of composite primary key)
This is a basic parent/child composition relationship between these two tables. Using adf business components, I have set up a master/detail relationship between these two tables. Of course they're connected via a view link which is based on an association which joins the two tables by application_name. I have a requirement to provide the user the ability to change, via a jsp within the application, the application name. This is accomplished by dragging the master view (Application) onto the jsp as a form. When they change the application name on the master view, the framework handles the automatic update of this same attribute (application_name) on the detail view. Works great. If i change my tables to use surrogate keys, they would look something like this:
Application (table's name)
application_id (primary key to be generated by sequence/trigger)
application_name
Menu (table name)
menu_id (primary key to be generated by sequence/trigger)
application_id (foreign key to application_id in Application table)
application_name
menu_name
Now, in this scenario, the association that joins these two tables will be based on application_id. This means that now if the user wants to change the application name (via a jsp in the application) in the master view, the framework won't automatically update application_name in the detail view. So I suppose I will have to programmatically ensure that if application_name of the Application table is changed, that new value is also 'cascaded' down to the application_name
in Menu table. How would I programmatically take the new application_name of the master and ensure that the application_name of the detail is updated as well. I suppose I'll have to control the post order of this too?
Also, this may be more of a db question, but if anyone has any strong opinions on whether the framework is more well suited to use surrogate keys instead of intelligent keys, I'd like to hear about it. I've found a couple situations in which it seems that using the framework with surrogate keys has an advantage over using intelligent keys. Thanks.

In your second scenario, can you clarify whether it is by design that you'd be storing a redundant copy of the application_name in the menu table? normally to display the application_name you could just include the master application entity usage as a reference in a detail view object based on the table without the redundant storage of the name. That way, you'd only have one place to update.
It's definitely possible to implement a cascade update like you're asking about, but it's even easier to only store the application_name in one place.

How to create surrogate key in dimension without unique value

Hi, I have a dimension where there is no column with unique value. I want to add a surrogate key to replace the existing primary key which is derived from concatenating 3 columns(e.g. 'A'||'B'||'C'). I'm thinking of using sequence. But this won't allow me to link the dimension to fact table. How do I come up with surrogate key under this situation? Thanks. ~Tracy

I'm actually trying to accomplish something similar myself.
In my sources I've got two sorts of customers, ones that are directly reported, and ones whose information is provided with sales records (this is stored in module ODS).
Of course identification is different, but in the datamart (module DWH) I'm sort of forced to use an equivalent way of loading (due to the way it first used to work). To accelerate lookups on dimensions, I copy the ODS surrogate key to DWH dimensions, but this does not work for the 'inbuilt' customers because they do not have a surrogate key in the ODS.
They DO have means of unique identification, and at first I thought I could concatenate these (also 3) columns to use as identification code. Unfortunately this is VARCHAR2, where the surrogate key is (naturally) NUMBER.
So now it looks like I'm forced to first build a table in ODS especially for these 'inbuilt' customers and assign a surrogate key (by sequence) to it, this way it conforms to how 'normal' customers are loaded into DWH.
I guess you'll have to pull of the same trick, i.e. create a table with either only the 'translation' of D-code to a surrogate key or all information that is fed into the dimension, which then can be used as a lookup or as complete source when loading data into your datamart.
Good luck, Patrick

How we use Surrogate Keys for snowflake dimension

Hi All,
my question is - How we use Surrogate Keys for snowflake dimension
i heard from some body Surrogate Keys only work with star schema.
please correct me if i wrong.
Regards,
Manish

Hi manishcal16PPS,
According to your description, you can only create natural key in your dimension. But it's not working when using surrogate key. Right?
In Analysis Services, the snowflake schema of the dimensions are represented by more than one dimension table in other words its takes multiple dimension tables to define a dimension. Surrogate key are just some extra, redundant, unique key based on the
natural key. So there's no direct relationship or some limitations between surrogate keys and snowflake schema.
In this scenario, since there's relationship between the two dimensions, you should create natural key. For using natural key or surrogate key. Please refer to an article below:
Surrogate Key vs. Natural Key
For understanding star/snowflake schema, please see:
Understanding Star and Snowflake Schemas
Regards,
Simon Hou
TechNet Community Support

Hash partitioning v. list partitioning on surrogate key + partition pruning

Hi,
Have a large fact table with surrogate keys, therefore queries are of form
select dimension.attribute..
from fact, dima, dimb..
where facta.dima_surrogate_key = dima.dimension_key
and facta.dimb_surrogate_key = dimb.dimension_key
and dima.attribute = <value>
and dimb.attribute = <value>
Would ideally like partition pruning to happen but will this happen if hash partition on facta.surrogate_key
Likewise could list partition on facta.dima_surrogate_key and further sub-partition on hash of factb.dima_surrogate_key.
Any advice much appreciated.

user5716448 wrote:
Hi,
Version 11.2.0.1
fact table structure
PRODUCT_ID NUMBER
RETAILER_ID NUMBER
OUTLET_ID NUMBER
CALENDAR_ID NUMBER
BRANCH_ID NUMBER
PUBLISHER_ID NUMBER
DISTRIBUTOR_ID NUMBER
TRANS_TYPE_ID NUMBER
TRANS_QTY NUMBER (10)
TRANS_VALUE (10,4)
No date on fact table (just surrogate_id for calendar whihc links to calendar/date dimension.
Although queries can be by date of transaction, most aren't.
Potential to grow to 3 billion rows.
Considering hash partitioning on the product_id, simply to break data down and product_id is the largest dimension.About hash partitioning – in this case it is probably all about the ability to run in parallel. Do not have any info on that, so I cannot comment further.
>
sqls are varied, lots of different types some query all dimensions, sometimes a few. Not the straightforward date examples in the manual.You can pick a dimension that is frequently used by the SQLs. I understand that there is no perfect one, but even if you pick just a “good” one you might have a good deal of partition elimination.
>
Users run 3rd part ad-hoc reporting layer which has to allow them to report against the star in any way they want.
Star transformation hint enabled. Have heard in deciding number of hash partitions, partition size should geneerally be < 2gb.
e.g transactions for a given product for customers belonging to a given multiple in a given week
select trans_qty, trans_value, m.prod_name, m.prod_num, r.cust_name, w.branch_name, rtt.trans_date, rtt,trans_type
from retailer_transaction rt, media m, wholesaler w, calendar c, retailer r, trans_type rtt
where rt.issue_id = m.dimension_key
and m.prod_num = 600
and rt.branch_id = w.dimension_key
and rt.outlet_id = r.dimension_key
and r.multiple_num = 700
and rt.calendar_id = calendar.dimension_key
and m.issue_year_week = 201110
and rt.trans_type_id = rtt.dimension_keyLastly, you need to focus on weather and how to partition your indexes (I assume you have bunch of bitmaps). This decision is at least as important as partitioning the table.

Mapping in OWB with primary key and foreign key relationship

Hi all,
I am new to this datawarehousing field. I have just started my career. I have to now create a mapping in owb where a table has a field which is a primary key of another table in the same staging area. If you guys could help me out with the a method it can be created that would be very helpful to me.
I thought of 2 ideas,
1. If I can use a look up, but then I am not sure if i can use a lookup for primary key, foreign key relationship. If I can use also, I do not know how to use that.
2. What if I can directly take that the first table and link the primary key of that table to the second table which uses that primary key of the first table as one of its fields.
I do not know how feasible these methods are. Please guys help me out.
Thanks in advance.

I have a similar case where table a and table b having relation but table a got inserted with data and table b is empty so there no values for foriegn key column in table b to realte with table a.
Now i want to load table b foriegn key with primary key column values of table a.
how can we do this in owb
thanks
kumar

Replacing natural keys with surrogate keys

Similar Messages

Maybe you are looking for