Data Analysis

Data analysis is nothing but understanding the contents of the data such as structured data like columns, tables and draw a business conclusion from the available data. Data Analysis needs to play a key role in current delivery model. It directs the requirements from business standpoint and helps developers to create a base point for understanding the requirement and providing the solution in best way possible. Nobody in earlier days thought that the content in the data can decide the performance of the system until parallel processing systems came into the picture.

Most of the product, developments are happening without proper understanding of data which in turn is making the system inconsistent. To make the most robust and viable product “KYD” should be in the mainline rather be at the back end step to have a most feasible and generic product while designing frameworks for bigger data sets.

Data Analysis should be added as a key stage in the software development life cycle these days. It not only reduces the loss incurred in the product developed a great extent but also helps in building a robust product. As a part currently data analysis is coming as a minor part of development. It should be coming as a separate stage of software development life cycle as documents such as data profiling, data model, Data Mapping can be accurately done based on “KYD” ( Know Your Data ) Principle.

Business stand point:

Data Analysis helps business to better project the requirements in the way the business actually want to. With the examples handpicked it better make the audience what actually business want the team for accomplishment.

Developer standpoint:

In development base data analysis plays a key role in understanding the requirement in clear visuals, designing the requirement and implementing the solution with the same visuals business is expecting.It also helps in the distribution of data with less skew among the nodes of the system which in turn increases the performance of the system. Thereby it reduces the cost of maintaining the system.The developer base should definitely understand the data they are working on and hence a separate phase should be allocated to the developers in the entire life cycle predominantly the early stages of the SDLC. This helps to make every person in the sprint to be on the same page. This will help the requirement not only complete with full phase with confidence but also helps in gaining the cost of implementing the solution , Tuning the solution and maintaining the solution. It helps in increasing the robustness of the framework.

Use case:

In a big-data system, When a developer is starting a task from initial stage of implementation. Nowadays people start with debugging the requirement, develop the code, and implement the code.How ever after this, we face lot amount of hurdles from both the business and at platform side. Considering this proper data profiling, data model and data mapping is prepared lot of hurdles can easily be resolved which will make the company gain lots of rupees.