Complicated stuff first — Cube is a method of storing data in a multidimensional form, generally for reporting purposes. In cubes, data (measures) are categorized by dimensions. Cubes are often pre-summarized across dimensions to drastically improve query time over relational databases.
Simple stuff — Cube is a data storage mechanism beyond rows and columns.
Eg. Sales of a particular product can be stored in a transaction table with sales figures but sales figures across years and across various cities cannot be graphically or physically represented just from one table.
We need various dimensions that may be related to the fact tables. In this case then the fact table would be the sales table and the dimensions would be time and city tables.
Let me give you some classic example of sales and marketing methodologies and the factors that may determine sales.
- For marketing of a new brand of soap there are a lot of things that are taken into conditions. this may include the following. Current preference, price, affordability of people, religious beliefs, color of soap, political standing, political inclination, political influence, age group, gender ratio, habits, water availability, distributor network, seller network, population and other market research activities. All these things cannot be clubbed into one entity for quick reference. In this case the dimensions are identified and they are associated with facts to give a clearer view for the product to the assessment team to come to the conclusion of marketing and sales strategy to be applied in this place.
- Another example would be two-wheeler sales. Road conditions, public transport availability and frequency, work force location, work location, fuel cost and efficiency, product capacity and other such related factors determine sales numbers.
- One more example could be Gym membership. When a physically fit looking neighbor dies of a heart attack then rest of the locality gets into workout and exercise groove. The factor influencing the increase in gym membership are not only marketing or pricing. but there can be other influencing factors as well.
In all the above examples we were trying to gauge various datapoints from a single view. To analyse these data points from a single view needs the requirement of a cube that can save data from various sources of different dimension in a group to render this data graphically to achieve business objectives.
In simple words, cube is a multidimensional table with all the relevant fact and dimension details.
Post comments if this helps
