Sithu

• Oct 4th, 2005

 

Type-1

Most Recent Value

Type-2(full History)

i) Version Number

ii) Flag

iii) Date

Type-3

Current and one Perivies value

sai

• Oct 7th, 2005

 

fine

Henry

• Oct 18th, 2005

 

SCD

Type 1 : overwrite data is to be there.

Type 2 : current, recent and history data should be there.

Type 3: current and recent data should be there.

sranjanim

• Oct 21st, 2005

 

Henry Wrote:

SCD

Type 1 : overwrite data is to be there.

Type 2 : current, recent and history data should be there.

Type 3: current and recent data should be there.

ravi kumar guturi

• Mar 14th, 2006

 

SCD means If the data in the dimension is happen to change very rarely,

Mainly SCD 3 types

1)SCD-1:HERE THE PREVIOUS DATA OVERWRITE BY CURRENT DATA

MEANS HERE ONLY MAINTAIN CURRENT DATA.

2)SCD-2: HERE JUST ADD THE ADDITIONAL RECORDS

IN SCD2 3 TYPES1)VERSIONING2)FLAGVALUE3)EFFECT DATE RANGE

VERSIONING: MEANS HERE JUST SEND THE UPDATED RECORDS TO THE TARGET ALONG WITH VERSION NUMBER.

NEW RECORDS WILL BE SENDING TO THE TARGET ALONG WITH PRIMARY KEY

FLAGVALUE:HERE UPDATED RECORDS SEND TO THE TARGET ALONG WITH 0

AND RECENT RECORDS SEND TO THE TARGET ALONG WITH 1

EFFECTIVE DATE RANGE:MEANS HERE TRACKS THE BOTH PREVIOUS AND CURRENT DATA

3)SCD-3: HERE MAINTAINS JUST PREVIOUS AND CURRENT DATA.

Arvind Kumar

• Jan 29th, 2007

 

Type1: Rows containing changes to existing dimensions are updated in the target by overwriting the existing dimension. In the Type 1 Dimension mapping, all rows contain current dimension data. Use the Type 1 Dimension mapping to update a slowly changing dimension table when you do not need to keep any previous versions of dimensions in the table. Type 2: The Type 2 Dimension Data mapping inserts both new and changed dimensions into the target. Changes are tracked in the target table by versioning the primary key and creating a version number for each dimension in the table. Use the Type 2 Dimension/Version Data mapping to update a slowly changing dimension table when you want to keep a full history of dimension data in the table. Version numbers and versioned primary keys track the order of changes to each dimension. Type 3: The Type 3 Dimension mapping filters source rows based on user-defined comparisons and inserts only those found to be new dimensions to the target. Rows containing changes to existing dimensions are updated in the target.

desaial

 Profile Answers by desaial 

• Dec 9th, 2008

 

SCD means Slowly Changing Dimension.

The price of the product, address of the person, name of the city are few examples of SCD.

There are three methods to capture SCD.

Type 1: Overwriting the old values

As the old value is overwritten by the new (current) value, there is no way to find out what the old (historical) value(s) were.  In other words, it holds ONLY current value.

Type 2: Adding an additional record in the dimension table

In this case, an additional record is added in the dimension table.  This method is very useful for the ACCURATE reporting purpose as all the OLD VALUES are captured.  In this method, we need to change the table structure so that PK violation doesn't take place.  Generally speaking, addition of the columns like "Effective Date" and "Expiary Date" suffice the need.

Type 3: Addition of the column (old value) in the dimension table.

In this method, when the value of the dimension is updated (by the current value), the old value column of the table is updated by the most previous value of the dimension.  Thus we have the CURRENT value and the MOST PREVIOUS value of the dimension.  This method is also NOT EFFECTIVE when reporting is needed over the period of time when the value of the dimension has changed more than once.   

Thanks.

Alpesh

ypinky

 Profile Answers by ypinky 

• Jul 26th, 2010

 

SCD-1 (Current)

Overwriting the old values in the dimension record.

 

SCD-2 (History and Current)

Creating an additional dimension record at the time of the change with the new attribute values.

 

SCD-3 (Previous and Current)

Creating new current fields and move the old attribute in a precedent field.