Hi All,
I want to create a cube using MSSQL2000 i dont know any thing how to
create and what is parameter required so if any body help me i will be
thankfull to him/her.For information : i have 4 tables
custdetailtable which contain all about a customer as per example
customer ID,name,add,CUSTOMERID IS PRIMARYKEY,
Proddetail table which contain all about a product mean
productcode,productname,productvalue,dman (dateof manufacture),dexp
date of expire,etc etc PRODUCTCODE IS PRIMARY.
Sales this table contain data aboutsale mean sales detail
cusomerid,productcode,noofpieces,priceperunit,prod uct name,date of
sale. etc etc SALEID IS PRIMARYKEY,PRODUCTCODE AND CUSTOMERID IS
FOREIGNKEY AND PRODUCTCODE IS PRIMARYKEY,
and atlast i have a table name stok detail which contain product
id,DTOFman and DTofexpire and amount of product
here product id,DTOFman and DTofexpire are foreign key.

Hi

I am not an expert on this but I would expect a good start is to read:

http://msdn.microsoft.com/library/de...build_6b77.asp
http://msdn.microsoft.com/library/de...build_0lyq.asp

I would also recommend reading some of the background information.

John

"Biswajit Barik" <biswajitbarik (AT) rediffmail (DOT) com> wrote

It is impossible to pass full data warehousing knowledge in a newsgroup
posting but here is my attempt

First get it out of your transactional environment (OLTP). Use DTS to
transfer this to another database (preferably another server depending
on size).

You will want to decide the subject matter of your cube which for you
would be the sales. This is your lowest granularity and you will use
this table to create what is called your fact table. Your fact table
will consist of the lowest granular aggregations (sales) and foreign
keys back to the other tables which will be used as dimension tables.

This is the structure that you will use to build the cube from. You need
to install Analysis Services for SQL Server 2000. Go into Analysis
Manager and say new database. call it whatever you want.Here you will
add a data source that points to wherever you sent your transformed
table. then right-click on cubes and say new cube, use the editor not
the wizard. it will ask you what tables you want to use, specify your
fact table (the one with sales data and your key columns) and add it.
You can then add your dimension tables (details - like customer, product
and possibly stock in your case). In the cube editor you will create
dimensions (all real cubes need a date dimension) dimensions are
typically what the customer says they want to see the data "by"
Examples:

I want to see the number of sales by region (region would be a
dimension)
I want to see the average sales by customer (customer would be a
dimension)

Again, these are examples, might not apply to you.

Then you will pick your measures, these are your aggregations, WHAT do
they want to see (dollars, total orders, etc.)

Calculated members are more advanced and you can get into those by
looking at MDX examples. From there you need to process the cube and
then you can connect to it by using Excel pivot tables (or something
better once you get the hang of it)

If you want to know the theory and use of data warehousing/cubing get
Ralph Kimball's data warehouse lifecycle toolkit book, if you want to
quickly learn how to use analysis services get Reed Jacobson's "Analysis
Services: Step by Step"

HTH

Ray Higdon MCSE, MCDBA, CCNA

*** Sent via Developersdex http://www.developersdex.com ***
Don't just participate in USENET...get rewarded for it!

BTW: Before I get attacked , my post, of course, if no where near full
data warehousing information. There are several things that were left
out, checking out those two books and websites like dmreview.com or
intelligententerprise.com will help get more DW knowledge.

HTH

Ray Higdon MCSE, MCDBA, CCNA

*** Sent via Developersdex http://www.developersdex.com ***
Don't just participate in USENET...get rewarded for it!