A New Approach to DW/BI

 

Aside from obvious advantages like speed and efficiency, 1010data provides important benefits with respect to the way things get done. Some of these apply to database design and development, others to analysis and reporting, but all are significant enough to constitute a true paradigm shift.

Database Design and Management

The speed of our system obviates the need for data marts, OLAP cubes, indexing and key-field definition, and, more significantly, allow the most granular data to be stored and used by applications and users. Even raw data can be loaded into the database and used directly by applications without loss of performance. This has important implications for what analysis can be done (see the next section) and also for the development process.

Normally database implementation requires extensive requirements gathering and design efforts resulting in an ETL process that transforms raw data into "clean", standardized information. The design process must also define key fields and indexes, all of which takes time and effort. Because the ETL process is what feeds the database to begin with, the database cannot be instantiated until the ETL process is designed and implemented. This means that every application must wait until the ETL process is completed for all the data, even though a given application may use only a tiny subset of the data.

With 1010data's fast database technology, lightly processed or even raw data can be loaded into the database and there is no need for indexing or defining keys or other relationships. The database can therefore be instantiated and made available to application developers and users almost immediately. Of course the data has to be cleaned before it is used in analyses and reports, but the implementation of a given application or analysis can include cleaning the relevant data. There is no need to wait until all the data is cleaned. This can cut the time to delivery for the application from months to weeks or even days.

Loading raw data is also the ultimate way to do data exploration. Why settle for data-exploration tools that provide limited insight into raw data when you can see all the data in our UI and apply the full analytical power of our back end to it?

Analysis and Reporting

With access to the most granular (or even raw) data, users are unconstrained in their analysis. Add our user-friendly UI and application developers and even end users can quickly create sophisticated analyses and reports. The result is an unprecedented access to data that has tremendous value to a customer.

Further, any table in our system, even one uploaded by end users through our UI, can be joined to any other table(s). So not only can users define their own data relationships but they can add their own data as well.