When do we consider snowflaking a star schema? - etl

In the below dimensional model, all the dimension table follows the star schema except the prod table, which is snowflaked and has prod_sub table as a child table. What is the strategical point to consider snowflaking a star schema?
Sales Table - Fact Table
Prod_id (fk)
Promo_id (fk)
store_id (fk)
Promo Table - Dimension Table
promo_id (pk)
Store Table - Dimension Table
store_id (pk)
Product Table - Dimension Table
Prod_id (pk)
prod_class_id (FK)
Prod_name
Prod_brand
Prod_sub Table- Dimension Table (snow flaked)
prod_class_id,
prod_class
prod_sub_class
prod_family

According to Kimball : Star Schema is the best way of designing a data model for reporting, You will get the best performance and also flexibility using such a model.
So no need to snowflake it.
Then your Product Table Dimension Table becomes :
Product Table - Dimension Table
Prod_id (pk)
Prod_name
Prod_brand
prod_class
prod_sub_class
prod_family

Related

Laravel 5.2 - Can You Create A Relationship Between A Pivot Table And A Table It's Not Pivoting For?

I have a pivot table called user_store. It's used to establish a many to many relationship between the users table and the stores table and looks something like this:
user_id - int
store_id - int
user_age_range - string
I have another table called user_memberships. It's columns look like:
user_id - int
store_id - int
membership_cost - decimal
membership_expiration - date
I need to establish a relationship between the user_memberships table and the user_store pivot table using the user_id and store_id columns present in both tables.
How would I do this when the user_store table does not have its own model (since it's a pivot table)?

Create table as select statement primary key in oracle

Is it possible to specify which is the primary key on creating table as select statement? My aim is to include the declaration of primary key on the create table not modifying the table after the creation.
CREATE TABLE suppliers
AS (SELECT company_id, address, city, state, zip
FROM companies
WHERE company_id < 5000);
Yes, it's possible. You would need to specify columns explicitly:
CREATE TABLE suppliers (
company_id primary key,
address,
city,
state,
zip
)
AS
SELECT company_id, address, city, state, zip
FROM companies
WHERE company_id < 5000;
Here is a demo
Note: in this case primary key constraint will be given a system-generated name. If you want it to have a custom name you'd have to execute alter table suppliers add constraint <<custom constraint name>> primary key(<<primary_key_column_name>>) after executing(without primary key specified) CREATE TABLE suppliers.. DDL statement.
Yes, it's possible.You can try referring below example.
create table student (rollno ,student_name,score , constraint pk_student primary key(rollno,student_name))
as
select empno,ename,sal
from emp;
You can create Suppliers table explicitly and if any column have primary key in companies table, then copy that structure from companies table.
Or
Create a same column defining primary key that you want and copy that column from companies table!

MODIFY or ADD to add NOT NULL constraint to a column? Oracle sql

ORDERS table in the Oracle Database:
ORDERS
ORDER_ID NOT NULL NUMBER(4)
ORDATE_DATE DATE
CUSTOMER_ID NUMBER(3)
ORDER_TOTAL NUMBER(7,2)
The ORDERS table contains data and all orders have been assigned a customer ID. I'm trying to add a NOT NULL constraint to the CUSTOMER_ID column. Would I use MODIFY CONSTRAINT or ADD CONSTRAINT? I was told you have to drop the constraint and ADD the new one, but if there is no existing constraint to Customer ID number, would it be MODIFY?
alter table orders modify customer_id not null;
Just MODIFY the column:
alter table orders modify customer_id not null;
Alternatively, you could add an [overkill] constraint in the form:
alter table orders add constraint nn1 check (customer_id is not null);
Just use the first form.
As a side note, some databases (such as Oracle) consider those two constraint different and somewhat separate: the former is a column constraint, while the latter is a table constraint. Oracle keeps track in case you drop one, while the other is still in effect.

Create a generic DB table

I am having multiple products and each of them are having there own Product table and Value table. Now I have to create a generic screen to validate those product and I don't want to create validated table for each Product. I want to create a generic table which will have all the Products details and one extra column called ProductIdentifier. but the problem is that here in this generic table I may end up putting millions of records and while fetching the data it will take time.
Is there any other better solution???
"Millions of records" sounds like a VLDB problem. I'd put the data into a partitioned table:
CREATE TABLE myproducts (
productIdentifier NUMBER,
value1 VARCHAR2(30),
value2 DATE
) PARTITION BY LIST (productIdentifier)
( PARTITION p1 VALUES (1),
PARTITION p2 VALUES (2),
PARTITION p5to9 VALUES (5,6,7,8,9)
);
For queries that are dealing with only one product, specify the partition:
SELECT * FROM myproducts PARTITION FOR (9);
For your general report, just omit the partition and you get all numbers:
SELECT * FROM myproducts;
Documentation is here:
https://docs.oracle.com/en/database/oracle/oracle-database/12.2/vldbg/toc.htm

Copy entire values of 2 columns from one table to another ensuring the relationship

I have a table STUDENT with columns st_id,name,age,dept_name. Now I want to create a new table STUDENT_DESC with columns st_id,dept_name,st_desc. So I need to copy all the values of st_id and dept_name to the newly created table STUDENT_DESC. I need to ensure relationship while copying st_id and dept_name , the dept_name should be corresponding to st_id.So how can I do it in PL/SQL?
insert into STUDENT_DESC (select st_id, dept_name, null from student);
this will simply copy all the records. The third column st_desc is left empty (null)
To ensure referential integrity you would add a primary key and a referential integrity constraint to the STUDENT_DESC table
However, note that in many cases it could be "wrong" to introduce a second table containing student data like that. It could be "better" to add st_desc to the STUDENT table.
I'm not sure I understand your data model, but at face value you can create your table simply:
CREATE TABLE student_desc AS SELECT st_id, dept_name FROM student;
ALTER TABLE student_desc ADD (st_desc VARCHAR2(..));
Fill in the .. with the desired max size for st_desc.

Resources