How Data ware house can help you make better decisions A beginners guide

The beginning
What is a warehouse?
Warehouse is collection of huge amount of periodical data that is used to study the behavior of the business over the period of time. 
Let’s assume that we have stored the data about a sale of products in different regions over a period of say 10 years from 2005 to 2015.
With Warehouse, being implemented perfectly, we can easily identify:
        1. Which product has been sold maximum in the month of September 2011?
        2. Which product has achieved maximum profit margins in the 3rd quarter of the financial year 2010?
      3. What is the trend for sale of laptops in Metropolitan city in 2nd quarter of year 2010 as compared to 2nd quarter of last 3 years?
 
So in simple terms, warehouse is a platform where we store:
        1. Raw data from various sources.
        2. Summarized data generated using this raw data.
        3. Data about the data called, metadata, to understand
       - What is the storage requirement of the data?
       - What is the nature of this data? Whether it is number or character or a date? 
 
Life cycle of a Warehouse

Image source- https://goo.gl/MfsVlV
 
Life cycle of a warehouse is all out the journey of data through it. It accepts raw data as an input and stores it as a set of meaningful information. Let’s go through all of the steps one by one. 
Initially, this raw data, having no meaning, is required to be collected from different possible sources.
        1. Uniformity is required to be imposed on the data collected from those different sources, if required.
        2. This uniformed data is now required to be transformed and scheduled to get loaded into the warehouse.
        3. This transformed data is further processed to do analysis.
        4. This analyzed information is then passed to the reporting systems to generate dynamic and meaningful reports.
 
Warehouse Architecture
Over the period of time, many theories have been put up regarding the architecture of a data warehouse. I could come up with 3 different ways that includes majority of patterns in warehouses.
        1. Centralized architecture.
        2. Federated architecture.
        3. Tiered architecture.
However, before explaining these types, I do feel it is important to have few words about data marts. Data marts are a portion of a data warehouse that contains data about a specific group. For example, if we talk about a sales details of an organization, then we can have data marts based on the territory, financial year, product or even sales representative wise. I hope data part is clear to all you.
So now, a stage is set where we can start with these architecture types.
Centralized architecture: In this type, we have single physical data Warehouse. So loading a data in warehouse is simple. However, accessing it is time consuming. This is the most traditional way of implementing a warehouse. 
Federated architecture: In this type, we have no physical data Warehouse. However, we have a logical area, that collects data from various sources and load it into physical data marts. This type is pretty useful, when the end user have all their analysis to be done category wise.
Tiered architecture: In this type, we have   physical data warehouse as Well as physical marts. Clearly with this approach, we can cover various varieties of requirements with faster access. However maintaining a change is a big challenge.
 
Conclusion
Warehouse is a big database that accepts raw data from various sources, processes it and store it in a specific format that helps us in analyzing the data with sound precision.
I hope this article helped you in understanding basics of warehouse. In future, we will be posting in depth discussion about these aspects.

Hello friends!!! Decision making is the one of the key aspects of any business for being successful. I am sure that all of you will agree with me that for any kind of decision making, we need data, a huge amount of data. Gone were the days when we relied on few GBs of data to come up to any conclusion. Today, we need Yottabytes of data for the same. However, to have this much ample amount of data with us, we need to have a special storage system. So in this blog, we will discuss about this special storage system called warehouse. We will also discuss about various terminologies that revolves around it.