Discussion Locked

Data WareHousing Selection Criteria

0 helpful
4 replies
Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.

Hi
I am working in a Mobile Telecommunication industry as a Senior Systems Analyst
Currently we are in phase of acquiring a data warehouse project.
Due more poor knowledge of such project, can you help in finding the best practice or selection criteria for the following?
-Data Warehouse Architecture
-Data Modeling
-ETL (Extract, Transform, Load) tools

Thanks

4 replies

by 20 + years

You may want to pick up a copy of "Data Warehouse Toolkit: Practical Techniques for Building Dimensional Data Warehouses (Includes CD-Rom) by Ralph Kimball".

This gentlemen is credited with designing the data warehouse concept. It will get you started on understanding the architecture. For data modeling I use System Architect but can't stand it. I prefer to use "ErWin" it is a much better tool. You can also look at the "Rational Rose" suite of tools. ETL tools run a broad spectrum and can cost very little to quite a bit. You need to have some ideas on the data you will be working with and the volumes. Informatica is a good one I have used in the past.

Data Warehouses are typically very complex and can be very expensive as well. You will require individuals who know the data requirements and the business processes you are trying to support. If you are truly new to this area, you may want to look at outsourcing. It may give a better chance at success. Good Luck

by sanjay_lists

Hi,

Data Warehousing topics on the internet is very dispersed. There is no one place where you can get all the info. I did most of my searching for my paper on data warehousing using google and www.datawarehousing.com. Interestingly for you, they have a paper this week on ETL tools which you have requested. I also have my paper on Data Warehousing and Decision support systems. If you want a copy of that pls. let me know you email address by sending an email to email2sanjay@yahoo.com, I will be glad to send you a copy.

Thanks,
Sanjay.

by gokul rajan

Hi,
After reviewing your question, I can suggest you presuming that you have got "Basic Database structure streamlined" you can adapt any kind of Data warehousing tools to kick start your project , in the initial stage.

Any kind of Data warehousing projects, starts from identifying & Analysing the data-points and data-inputs in your database model.

Try to understand where and how the data is being stored and streamlined into database. Study the scope of data modulation. Study the Inputs with criterias and sketch a rough diagram. Once you have the schematic diagram sketched, you can now implement Datawarehousing tools.

Data warehousing concepts are simple, Pull data from existing (streamlined) Database and use that for studying the Market, Consumption of products, Finance analysis, Market Research and so many components. Well the only thing you have to bear in mind is the DATA is REAL (24*7*365) and has to constantly monitored, unlike Data-Marts.

Considering the tools, Since you are "System Analyst whatever is present with your firm / corporate has to be followed. I can suggest you ETL Tools and few From ORACLE / Sybase. Just let me know if they are accepted.

Regards
Gokul R

by shanghai sam

Hi,
I will try to address it from a different end.

Datawarehousing also very much known as Business Intelligence and different vendor addressing it from different end. Oracle provides lots a tools scattered in many place in 9i like DBMS_OLAP package, Meterialized view, Insert statement to read exrternal file and many more.

If you have SQL Server 2000 and Analysis Service handy with you, install the demo of BI and you will get a hang of it. MS DTS (ETL) helps you to read from different source/format data and pump in to Warehouse DB or to your staging. You can programm your extraction by using DTS object strucuture (DATAPumptask2 etc.) but for basics GUI would be good enough to understand. ETL is very important in DW as that is the basic element to for your warehouse. Informatica is probably the best in this.

DW concepts are same in every platform, so you can get a good idea by reading article from MSDN.

Also, you could do the same in IBM DB2 UDB7.x and OLAP starter kit. It has a full functional demo which can be up & running in a very short time.

You can get enough material on data modelling to start with, in internet if you search by 'STAR Schema design'. DW mainly uses two type modelling as STAR schema, and Snow flake schema. In Star schema there is a fact table which holds summable data at the lowest level and is related with one or more Dimension table by foreign key relationship. Snow flake is further normalisation of the Dimensions. Every warehouse needs a staging area which is more or less like main DW schema but may have more table for pruning, converting and cleaning data for the warehouse.

Trying out demo will give you a strong induction to BI and after that you could use the tools as per your choice.

Thanks,

Debasis Ray

This discussion is closed to new replies.