Answer Posted / manoj
When developing a dimensional model, we often encounter miscellaneous flags and indicators. These flags do not logically belong to the core dimension tables.
A junk dimension is grouping of low cardinality flags and indicators. This junk dimension helps in avoiding cluttered design of data warehouse. Provides an easy way to access the dimensions from a single point of entry and improves the performance of sql queries.
Example: For example, assume that there are two dimension tables (gender and marital status). The data of these two tables are shown below:
Code:
Table: Gender
Id Gender_status
----------------
1 Male
2 Female
Table: Marital Status
Id Marital_Status
----------------
1 Single
2 Married
Here both the dimensions have low cardinality flags. This will cause maintenance of two tables and decrease performance of sql queries.
We can combine these two dimensions into a single table by cross joining and can maintain a single dimension table. The result of cross join is shown below:
Code:
id gender mrg_status
--------------------
1 Male Single
2 Male Married
3 Female Single
4 Female Married
This new dimension table is called a junk dimension. This will improve the manageability and improves the sql queries performance.
| Is This Answer Correct ? | 12 Yes | 0 No |
Post New Answer View All Answers
How many clustered indexes can you create for a table in dwh? In case of truncate and delete command what happens to table, which has unique id.
Where we use surrogate key explain with example?
Hi Friends, Can anyone tell me. is there any certification in Database testing or BI testing or ETL testing international certification? Thanks ArunKumar
What is the difference between snowflake and star schema?
What is is the guidelines to build universe with better performance? R performance tuning issues of universes?
Explain techniques of error handling - ignore , rejecting bad records to a flat file , loading the records and reviewing them (default values)?
Tell me what is full load & incremental or refresh load?
Explain piconet?
Explain difference between data ware house & data mart?
What is the differences between star and snowflake schemas?
What is freehand sql?
What is a view as it relates to system modeling?
Can you state some of the innovations throughout history?
How to improve session performance in aggregator transformation?
Explain what is power center repository?
